ADVERTIMENT. Lʼaccés als continguts dʼaquesta tesi queda condicionat a lʼacceptació de les condicions dʼúsestablertes per la següent llicència Creative Commons: http://cat.creativecommons.org/?page_id=184
ADVERTENCIA. El acceso a los contenidos de esta tesis queda condicionado a la aceptación de las condiciones de usoestablecidas por la siguiente licencia Creative Commons: http://es.creativecommons.org/blog/licencias/
WARNING. The access to the contents of this doctoral thesis it is limited to the acceptance of the use conditions setby the following Creative Commons license: https://creativecommons.org/licenses/?lang=en
GICOM (Composting Research Group)
DEPARTAMENT D' ENGINYERIA QUIMICA, BIOLÒGICA I AMBIENTAL
Escola d'Enginyeria
SUSTAINABLE CARBOHYDRASE PRODUCTION USING ORGANIC WASTES THROUGH
SOLID-STATE FERMENTATION: OPERATIONAL STRATEGIES AND MICROBIAL
COMMUNITIES ASSESSMENT
PhD Thesis
Alejandra Patricia Cerda Llanos
Supervised by:
Antoni Sánchez Ferrer
Teresa Gea Leiva
2017
Antoni Sánchez Ferrer i Teresa Gea Leiva, professors titulars del Departament d' Enginyeria Química,
Biològica i Ambiental de la Universitat Autònoma de Barcelona,
Certifican:
Que la enginyera bioquímica i màster en estudis ambientals Alejandra Patricia Cerda Llanos ha
realitzat sota la nostra direcció el treball amb títol "Sustainable carbohydrase production using organic
wastes through solid-state fermentation: Operational strategies and microbial communities
assessment", que es presenta en aquesta memòria i que constitueix la seva tesi per optar al Grau de
Doctor per la Universitat Autònoma de Barcelona.
I perquè en prengueu coneixement i consti els efectes oportuns, es presenta a l'Escola de Enginyeria
de la Universitat Autònoma de Barcelona l'esmentada tesi, signant el present certificat.
Dr. Antoni Sánchez Ferrer Dra Teresa Gea Leiva
Bellaterra, Febrer 2017.
To my mother.
"Real education must ultimately be limited to men/women who insist on knowing, the rest is mere
sheep-herding"- Ezra Pound.
"Sólo con el corazón se puede ver bien, lo esencial es invisible para los ojos"- Antoine de Saint-
Exupèry.
Acknowledgements
This thesis was carried out thanks to the financial support of the Spanish Ministry of Economy and
Competitiveness for the "Waste to product" CTM2012-33663 and BIOPRO project, the European
commission for the "Decisive" CTM2015-69513-R project and to the Universitat Autònoma de
Barcelona for the predoctoral fellowship granted (PIF-UAB).
In first place, i have to thank both of mi supervisors: Tere and Toni. Thanks for providing me with the
great opportunity to join the big GICOM family and for all of the support, trust and patience (above
all patience!!!). You two are an endless source of inspiration for me in both, personal and professional
aspects. I have to specially acknowledge the coolest, craziest and nicest supervisor i could ever ask
for: Tereeee!!!. More than a supervisor, a great person and friend and most definitely one of the
persons that i will remember with a smile whenever i remember my stayed at UAB.
To all the GICOM group, thanks for all the shared moments even those including crying, stress,
sweat, sludge and terrible odours!!. To my only (PhD) padawan: Pedro, keep sharing love, energy and
happiness to the world, you made every minute spent in the "caseta" unforgettable, jajaja! I wish you
the best for the bright future ahead!. Ahmad, Nora, Maria (let's not forget the last name
Compostatge"), Cindy, Óscar, Eva, Dani, the grad padawans Laura and Aida (you don't have to
google it anymore)...and everyone who passed for the composting lab thanks for being there.
To my mexican BFF's: Gaby and CChan!. I love you both. We will definitely be seeing you guys!
even if we have to go to Cancún to stay for two month in your house CChan, ajjajaajaj don't forget!.
Gabriela, in spite of the toughest time you had the last couple of years you still keep your smile and
positiveness. I admire your will to carry on even when the odds do not favour you.
To my dearest office mates: Jordi and Xavi. Thanks for making our office the coolest place to work
(among others), to always put a smile of my face, to have the perfect taste in music ((each one on our
style but no matter what other people say, it is very important!) and for being there and for sharing a
piece of your lives with me. I will miss you both.
Last but not least, thanks to my dear family!!!! Without all of you we (Carlos, Vicente and me)
probably wouldn't have made it through all this time in Spain. To my parents: Vivi and Nelson, thank
you for shaping me the way i am. I'm deeply grateful for all you have given to me, especially my
beautiful sisters: Vivi, Coté and Gabi! I love you all forever and ever!. To "los suegros": Carlos and
Maggi, i couldn't ask for a better family!!! I love you and thanks for all of the love, support and trust!.
To Matias (Makikis) and Manuel (papum): I always wanted brothers and now i have two!!! I'm so
lucky!!!. We miss you all always, everyday.
To the loves of my life: Carlos and Vicente. I can't imagine my life without you both. Carlos, you are
the partner i always imagined and more. Thanks for holding my hand during this (sometimes difficult)
road, i love you!!. Vicentin!!, with a smile melts my heart and at the same time makes me strong.
With you i feel i can accomplish anything.
As the late Gustavo Cerati would say: Gracias Totales!!
Summary
According to the solid-state fermentation (SSF) research line of the composting research group
GICOM, the main goal of this thesis is to develop a sustainable process for the production of
carbohydrases using organic wastes in a representative scale. To achieve this goal, several organic
wastes were screened on its potential to produce cellulase at laboratory scale. Also, compost was
added as a complex mixture of biomass in order to provide cellulose-degrading microorganisms. After
the selection of the two best substrates: orange peels and coffee husk using 0.5 L reactors, the scale up
was assessed in 10 and 50 L reactors. All SSF were undertaken in near-adiabatic conditions using
thermally isolated reactors with a continuous monitoring of the biological activity reflected as specific
oxygen uptake rate (sOUR) and temperature.
The main challenges to overcome in SSF using organic wastes is on one hand the possibility to work
in a continuous configuration and, on the other hand, the reproducibility of the proposed process.
These two aspects were assessed in this thesis.
A first approach towards a continuous process was carried out for the production of amylase using
bread waste and soy fiber as substrate and adding Thermomyces sp as inoculum. Three different
strategies were developed in a sequential batch operation by removing part of the substrate in three
diffeerent stages of the process: during maximum sOUR, maximum amylase activity and maximum
amylase activity with a posterior enzymatic extraction. An increase of above 200% in amylase activity
production was obtained in all the assessed strategies.
Due to the success of the amylase production in a continuous configuration, the next step was to use
the operational conditions obtained at laboratory scale for cellulase production and develop a
sequential batch operation in 4.5L reactors. An initial compost addition as inoculum was considered
only at the beginning of the fermentation. After reaching the maximum cellulase production the
system started working in cycles, removing one part of the substrate and replacing it for fresh
substrate. Two substrate exchange ratios were evaluated: 90% and 50%. Using this configuration, a
continuous process was carried out, with a continuous production of cellulase and valorization of
organic wastes and eliminating the requirement for fresh inoculum in each cycle. The sequential batch
operation resulted in the successful production of cellulase with sustained values around 8-9 filter
paper units per grams of dry matter in both configurations. In addition. the microbial communities
present at the end of this operation was characterized, identifying several cellulose, lignin and
hemicellulose degraders. This fermented solid was described as a specialized inoculum able to
colonize the solid matrix of the reactor and provides the opportunity to develop a reproducible
process.
Finally, the reproducibility and consistency of the process was assessed in triplicates in a 50 L reactor
using the specialized inoculum and coffee husk as the substrate. In addition the gaseous emissions
were measured in order to fully understand the process and to determine the requirement of a gas
treatment unit. Results showed that the process in consistent and robust with minor emissions of
pollutants.
Resumen
De acuerdo con la línea de investigación de fermentación en estado sólido (FES) que actualmente se
desarrolla en el grupo de investigación de compostaje (GICOM) el objetivo principal de esta tesis es
desarrollar un proceso sustentable y a una escala representativa para la producción de celulasas
utilizando residuos orgánicos. Con el fin de lograr este objetivo se analizaron diferentes residuos
orgánicos y su potencial para la producción de celulasas a escala de laboratorio. Además, se utilizó
compost como inoculo mixto, con el objetivo de incorporar el matriz sólida microorganismos capaces
de degradar material lignocelulósicos. Luego de la selección de los mejores substratos en reactores de
0.5 L, se llevó a cabo el escalamiento del proceso en reactores de 10 L and 50 L. Todos los
experimentos fueron desarrollados en condiciones cercanas a adiabáticas utilizando reactores aislados
térmicamente. Durante el desarrollo del proceso, se realizó la monitorización de la actividad biológica
reflejada como la velocidad especifica de consumo de oxígeno (sVCO) y temperatura.
El principal desafío a superar cuando se trabaja con FES y residuos orgánicos son, por un lado, el
desarrollo de un proceso en continuo y, por otro lado, que el proceso desarrollado sea reproducible y
consistente. Ambas temáticas fueron abordadas durante el desarrollo de la tesis.
Un primer acercamiento a un proceso en continuo se desarrolló para la producción de amilasas
utilizando fibra de soja y residuos de pan como substrato y Thermomyces sp como inoculo. Se
evaluaron tres estrategias de operación en reactores secuenciales, retirando del reactor una parte del
sólido fermentado en el momento de: máxima sVCO y máxima actividad de amilasas. En esta última
se evaluó la reinoculación con y sin extracción enzimática. En todas las estrategias se obtuvo un
incremento por sobre el 200%.
Debido al éxito de la operación en continuo para la producción de amilasas, se desarrolló un proceso
similar para la producción de celulasas utilizando las condiciones operacionales desarrolladas a escala
laboratorio. Inicialmente se consideró la incorporación de compost como inóculo (10%) y cascarilla
de café como substrato (90%). Después de conseguir el máximo de actividad de celulasas, el sistema
se continuó operando en ciclos, substrayendo una cantidad determinada de material y reemplazándolo
por substrato fresco. Se trabajó con tasas de intercambio de 50% y 90%. Finalmente se pudo
desarrollar exitosamente un proceso en continuo para la producción de celulasas, logrando
producciones sostenidas en el tiempo entre 8-9 unidades de actividad por gramos de materia seca.
Adicionalmente, se caracterizó el sólido fermentado obtenido al final de la fermentación con la
finalidad de identificar las poblaciones microbianas presentes. En este sólido se identificaron una
amplia gama de bacterias y hongos capaces de degradar material lignocelulósico, por lo que se
catalogó como un inóculo especializado que podría ser utilizado con la finalidad de desarrollar un
proceso reproducible; sin necesidad de la adición continua de nuevo inóculo.
Finalmente, se evaluaron a escala piloto (50 L) la reproducibilidad y consistencia del proceso
utilizando el inóculo especializado y cascarilla de café como substrato en triplicados. Adicionalmente,
se analizaron las emisiones gaseosas generadas durante el proceso con la finalidad de determinar los
posibles requerimientos de una unidad de tratamiento de gases. Los resultados obtenidos en estos
experimentos mostraron que tanto la dinámica del proceso como la producción enzimática fueron
consistentes y robustas. Además, el proceso puede ser catalogado como amigable con el
medioambiente, debido a la valorización de residuos orgánicos, así como también debido a los bajos
factores de emisión generados por el sistema.
Index
Chapter 1. Introduction...........................................................................................................................1
Chapter 2. Research objectives ............................................................................................................35
Chapter 3. Materials and Methods........................................................................................................39
Chapter 4. Cellulase production by solid-state fermentation using organic wastes at lab-scale and
scale-up..................................................................................................................................................61
Chapter 5. Towards a competitive solid-state fermentation for enzyme production...........................91
Chapter 6. Reproducibility assessment of cellulase production through solid-state
fermentation.........................................................................................................................................115
Chapter 7. Additional experiments.....................................................................................................137
Appendix..............................................................................................................................................155
Chapter 1.Introduction
1
CHAPTER 1. INTRODUCTION
Part of this chapter has been published in: El-Bakry, M., Abraham, J., Cerda, A., Barrena, R., Ponsá,
S., Gea, T., Sánchez, A. 2015. From wastes to high value added products: Novel aspects of SSF in the
production of enzymes. Critical Reviews in Environmental Science and Technology. 45(18), 1999-
2042.
Chapter 1.Introduction
2
Chapter 1.Introduction
3
1.1 Introduction: Towards a circular economy
In the last years, the worldwide solid waste production has been estimated to be near 10 billion
tonnes, half of which is produced by industrialized countries. However, the wastes production in non
developed countries has also increased during the last decades, with people moving from rural to
urban areas, thus increasing the rate of hazardous and industrial wastes (UNEP, 2015).
These solid wastes are generally disposed in controlled or, in the worst cases, uncontrolled landfills,
incinerated or destined for animal feed, representing great environmental problems with its inherent
financial losses. These associated costs are mainly related to the high cost of transport in the case of
landfill disposal, the high amounts of water and energy inputs for incineration and in the case of
animal feed, the obvious suitability of every waste for animal consumption (Liguori et al., 2013).
Therefore, there is an urgent need for the development of more sustainable programs of waste
management, with the main objective to reduce associated environmental and monetary costs. In this
context, a beneficial waste management is based on circular economy that strives to reduce the waste
generation and to exploit the renewable resources (Allesch & Brunner, 2014). One of the most
interesting alternatives relies on the biorefinery concept, which is defined as the sustainable
conversion of biomass into a range of marketable bio-based products (platform chemicals, additive
molecules for the chemical industry, biomaterials, etc.) and/or bioenergy (biofuels, electricity, heat,
etc.) (Jacquet et al., 2015).
In this context, lignocellulosic biomass is considered as the major promising renewable resource, due
to its great availability worldwide and its great potential to generate bio-based products and
bioenergy. Therefore, lignocellulosic wastes should be considered as “renewable resources” that can
be reused to generate valuable and marketable products, replacing the exhaustible fossil-based
resources (Jacques et al., 2015). In parallel, there has been an increasing interest for the reduction of
fossil fuel consumption and the development of new sustainable biofuels mainly obtained from this
lignocellulosic biomass (Liguori et al., 2013). There is such a great interest in renewable energy that it
has been reported that this type of energy supplied a 19.1% of the global energy request in 2013, and
furthermore, this contribution is continued to expand in the next years.
In summary, there are two main concerns nowadays, on one hand it is the necessity of replacing fossil
fuels for biofuels and, on the other hand, there is the new biorefinery concept for the sustainable
management of organic solid wastes. Considering all these aspects and in response to these two main
concerns, bioethanol appears as an interesting alternative. Bioethanol can be obtained using
lignocellulosic wastes as cheap and widely available substrates. Assessments on bioethanol
production have proven that the main cost associated to this process is cellulase production,
Chapter 1.Introduction
4
accounting up to 40% of total costs (Arora et al., 2015). It is for this reason that most of the current
efforts on these issues are focused on the optimization of cellulase production and the development of
novel and cost-effective technologies to allow it.
1.2 Lignocellulosic materials
Lignocellulose is the largest biomass feedstock on Earth and one of the most studied raw materials.
This material is the main components of the cell wall of plants, accounting up to 90% of its dry
weight. These types of materials include agricultural residues (e.g stoves, cobs, husks, straws),
industrial wastes (e.g sawdust, paper mills discards), urban and domestic solid wastes (sewage,
organic fraction of the solid wastes), among others (Martins et al., 2011).
The lignocellulosic feedstock generally consists of three main structures: cellulose, hemicellulose and
lignin. A scheme of the structure of the cell wall of plants is presented in Figure 1.1.
Figure 1.1. Model of molecular structure of the main constituents of lignocellulosic material. Source:
Rubin (2008).
Chapter 1.Introduction
5
In plant derivates, these components constitute the supportive skeleton of the plant body. Cellulose
forms fine fibers to constitute the cell wall network skeleton. Hemicellulose and lignin are the filler
and binder among the fibers. This entire structure contributes to tensile strength and it is responsible
for the chemical resistance, providing hydrophobic conditions.
1.2.1 Cellulose
Cellulose is the major component in lignocellulose, accounting between 35-50% of the total cell wall.
Chemically it is a linear polymeric chain formed by several D-glucose molecules bound together by β-
1,4-glycosidic bonds, which can be generally represented by the formula (C6H10O5)n. The number of
glucose unit ranges within hundreds and thousands.
All alternate D-glucose residues in the same cellulose chain are rotated 180º. One glucose unit is the
monomeric unit of cellulose and the dimer, cellobiose, is the repetitive structural unit of the cellulose.
This structure is polarized: there is a non reducing group at one of its ends and a reducing group at the
opposite end (Varrot et al., 2003).
Cellulose has several glucose units organized in long and unbranched chains caused by –CH2OH
groups alternating above and below the plane of rings. The lack of side chains lets cellulose molecules
form organized structures. Cellulose has both crystallized (higher packing density) and amorphous
(lower packing density) regions. Hydrogen bonds and van der Waals forces connect cellulose chains
in the crystallized region into structures named elementary fibril. Aggregates of these structures are
called microfibrils, which stacked together make up the fibrils, which gives the cellulose fibers (Rojas
et al., 2015) (Figure 1.2).
In these cellulose fibers, crystalline and amorphous regions alternate. The crystalline regions are very
cohesive, with rigid structure, formed by the parallel configuration of linear chains, which results in
the formation of intermolecular hydrogen bonds. This set-up contributes to the insolubility and low
reactivity of cellulose, making it more resistant to acid hydrolysis and modifying fiber elasticity. The
amorphous regions are formed by cellulose chains with weaker organization, being more accessible to
enzymes and susceptible to hydrolysis.
Chapter 1.Introduction
6
Figure 1.2. Plant cell wall structure and microfibril cross-section (strands of cellulose molecules
embedded in a matrix of hemicellulose and lignin). Source: Rojas et al. (2015).
1.2.2 Hemicellulose
Hemicellulose is the second most abundant renewable organic material on Earth, next to cellulose. In
the conversion of lignocellulosic biomass into biofuels, the utilization of hemicellulose as a by-
product is essential to make posible the biotransformation into value added products.
Hemicellulose is a plant-derived polysaccharide consisting of many sugars monomers of pentoses (D-
xyloses and D-arabinose), hexoses (D-glucose, D-galactose and L-rhamnose) and another uronic acids
as a branched chain connected to the base chain such as glucuronic acid and D-galacturonic acids.
This polysaccharide contains from 200 to 700 monosaccharides residues (Varrot et al., 2003).
Then, hemicellulose is defined as an heteropolymer, which classification depends on the type of the
constituent monomer. Hemicellulose can be classified into xyloglucans, xylans, mannans and β -
(1→3,1→4)-glucans. Also galactans, arabinans and arabinogalactans can be included in the
hemicellulose group, even though they do not share the equatorial β -(1→4)-linked backbone
structure. In this way, hemicellulose form and structure will depend on the source, either from fruity
or woody origin (Bidlack et al., 1992).The branched and amorphous structure of hemicellulose, in
addition to their low molecular weight, make these polysaccharides readily metabolizable (Li et al.,
2010). This is important, considering further cellulose hydrolysis, for which it is mandatory to remove
Chapter 1.Introduction
7
big amount of hemicellulose, thus improving accessibility of cellulases for its substrate (Agbor et al.,
2011).
1.2.3 Lignin
Lignin is probably the most complex and least characterized molecular group among the wood
components. It is a recalcitrant aromatic polymer, without defined repetitive units. Its precursors are
three p-hydroxycinnamyl alcohols or monolignols (p-coumaryl, coniferyl and sinapyl) and their
acylated forms (Martínez et al., 2008).
Although these precursors are phenolic compounds, the polymer is basically non-phenolic due to the
high frequency of ether linkages between the phenolic position (C4) and a side-chain (or aromatic
ring) carbon of the p-hydroxyphenyl propenoid precursors, strongly predominant in the growing
polymer. Unlike cellulose and hemicelluloses, the lignin polymerization mechanism (based on
resonant radical coupling) results in a complex three-dimensional network.
The main purpose of the lignin is to provide strength and water permeability to plants, but also to
protect them from pathogens. This protection prevents the action of microorganisms and hydrolytic
enzymes hindering further cellulose and hemicellulose hydrolysis (Martins et al., 2011).
1.3 Cellulases
Cellulases are a group of enzymes of great industrial importance, especially regarding bioethanol
production. These enzymes are considered as components of the enzymatic system for the hydrolysis
of plant cell wall. Cellulase is a complex group of enzymes comprised of several enzymatic activities,
each containing more similar enzymes of the same class. Furthermore, cellulolytic enzymes often
possess the ability to react on a variety of substrates.
1.3.1 Structure and function
Structurally, cellulases are modular enzymes composed of functionally discrete units, referred to as
modules or domains. In general, cellulases structure normally consists of one catalytic domain and
one carbohydrate binding module, which is usually joined to the catalytic domain by an often
glycosylated peptide (Davies & Henrissat, 1995).
Chapter 1.Introduction
8
The carbohydrate binding domain main role is to favor the enzyme to cellulose contact by increasing
at the same time the effective enzyme concentration and the time the enzyme will spend in the near
proximity of the substrate. The absence of carbohydrate binding domain, i.e enzyme with only the
main catalytic domain, still have the ability to join to cellulose; however with less affinity for the
substrate (Rabinovich et al., 2002).
All cellulolytic enzymes belong to the o-glycosyl hydrolases (EC 3.2.1), hydrolyzing glycosidic bonds
between two or more carbohydrates, or between carbohydrate and non-carbohydrate. Classically,
cellulases have been classified according to the substrate specificity in two groups: endo- and exo-
glucanases. A third class of enzymes working together in synergy with cellulase are β- glucosidase.
a) Endoglucanases (EC 3.2.1.4): This group of enzymes, also named endo-1,4-β-D-glucanases or
endocellulase, acts on a random sector on the amorphous cellulose chain generating different sized
oligosaccharides. The main objective of this group of enzymes is the reduction of the polymerization
degree of the cellulose chain, making it more available for further hydrolysis.
b) Exoglucanase (EC 3.2.1.91): This group of enzymes is also named exo-1,4-β-D glucanases,
cellobiohydrolases or exocellulases. These enzymes act from the end of the cellulose polysaccharide
chain, processing along the polymer chain while releasing cellobiose as a main product (Davies &
Henrissat, 1995). Small amounts of glucose and celotriose can be also released to the media in early
stages of hydrolysis. One of the main features of these enzymes is their ability to act on
microcrystalline cellulose (Teeri, 1997). This is of special relevance considering the structural
resemblance of microcristalline cellulose to a pretreated real lignocellulosic substrate (Percival-Zhang
et al., 2006).
c) β-glucosidase: This enzyme catalyzes the hydrolysis of short soluble oligosaccharides and
cellobiose into glucose. This reaction takes place on the liquid phase of the media, which is one
important difference when comparing with the action of endo and exoglucanases. Both enzymes act
on the surface of cellulose. It is for this reason that β -glucosidase is often not included in the main
cellulase group. A schematic figure on the full cellulase action is presented in Figure 1.3.
Final products of the full cellulose degradation are glucose and cellobiose. It is of great importance
the removal of these compounds, due to their significant non-competitive inhibitory effect on endo-
and exo-glucanases. It has been reported that β-glucosidase action is the main bottleneck for optimal
cellulose degradation (Gusakov & Sinitsyn, 1992).
Chapter 1.Introduction
9
Figure 1.3. Schematic diagram of the action of the three enzymatic groups of cellulases. Source:
Ratanakhanokchai & Sakka ( 2013).
Many microorganisms are able to produce the set of enzymes able to degrade cellulose, among them,
bacteria and fungi (Kuhad et al., 2016). These microorganisms can be either aerobic or anaerobic and
be able to produce cellulases under mesophillic and thermophillic conditions. Among them, the most
studied genera are Trichoderma, Aspergillus, Clostridium and Cellullomonas (El-Bakry et al., 2015).
1.3.2 Factors affecting cellulose hydrolysis
When working with lignocellulosic materials, there are several challenges to overcome in order to
fully hydrolyze this complex solid matrix. These can include: the origin of the enzymes, thermal
inactivation, adsorption or end product inhibition.
The effective enzymatic hydrolysis of cellulose largely depends on the origin of the enzymatic
complex, i.e from bacteria or fungi, and the mode of action of the cellulases. On one hand,
considering the origin of the enzymes, the enzymatic systems of fungi and bacteria present great
differences on their structure and on the characteristics of the produced enzymes (Kuhad et al., 2011).
On the other hand, considering the action of enzymes, endoglucanases action mainly takes place in
less ordered and inner regions of cellulose, thus contributing to the reduction of polymerization
degree. Also, exoglucanases hydrolyzes from the end chain, releasing cellobiose, which have almost
negligible effect on polymerization degree reduction. These joint actions can affect the enzymatic
hydrolysis on different substrates (Yang et al., 2011). In addition, the action of these enzymes and
Chapter 1.Introduction
10
potential synergisms between cellulases (or even hemicellulases, such as xylanase) can significantly
improve hydrolysis rates on complex substrates (Kostylev & Wilson, 2012). However, it has been
stated that authors tend to oversimplify the synergistic effect among cellulases and hemicellulases,
which is not a trivial matter or even a fully understood process. This is due to the fact that there is no
current evidence that synergism requires interaction between the synergizing cellulases, since
cellulases from unrelated organisms often show cross synergism (Berghem & Pettersson, 1973;
Kostylev & Wilson, 2012).
The complexity in degradation of lignocellulose results from the three dimensional structure of the
plant cell walls and the complex interactions between different components. Access to cellulose is
hampered by hemicellulose as well as lignin. Some authors have established that lignin poses the
greatest obstacle to cellulose degradation, with hemicellulose having a lesser impact (Van Dyk &
Pletschke, 2012). In order to access cellulose structure for hydrolysis, it is mandatory the removal of
lignin. Hydrolizates of this material has been proven to present detrimental effect in enzymatic
hydrolysis, enzyme recycling and will affect directly the enzymatic extraction in productive processes
(Brethauer & Wyman, 2010; Li et al., 2016c). Lignin adsorption is an irreversible and non-productive
process formed by hydrophobic interactions, electrostatic and hydrogen bonding interactions, which
leads to a continuous loss of enzymatic activity during the hydrolytic process (Pareek et al., 2013).
Many studies have been performed In order to reduce the adsorption of cellulases into lignin.
Chemical pretreatments such as steam explosion are commonly used for lignin solubilization;
however these treatments are expensive, energy consuming and environmentally hazardous (Li et al.,
2016c).
In contrast with the non-productive adsorption of cellulase into lignin hydrolyzates, there is
productive adsorption of cellulase into cellulose. The insoluble characteristics of cellulose make
enzyme/substrate adsorption mandatory to fulfil hydrolysis purposes. Furthermore, adsorption has
been found to be one of the important parameters that govern the enzymatic hydrolysis rate of
cellulose (Gusakov & Sinitsyn, 1992). It has been shown that cellulase adsorption is facilitated by the
carbohydrate binding modules, although, in some cases, catalytic domain can also specifically adsorb
to cellulose independently of carbohydrate binding modules. The efficiency of the adsorption process
varies with the source of cellulose and the pretreatment methods and it is correlated with the
crystallinity and the specific surface area of cellulose substrates. For practical use of cellulosic
materials, the cellulose structural properties and their effects on cellulase adsorption and the rate of
hydrolysis must be taken into consideration (Lee et al., 1982).
The enzymes of the cellulolytic complex may be subjected to catabolic repression by the final product
of hydrolysis. In this sense, β -glucosidase is responsible for preventing accumulation of cellobiose,
Chapter 1.Introduction
11
controlling the overall rate of the cellulolytic hydrolysis reaction, exhibiting a crucial effect on the
enzymatic degradation of cellulose (Leite et al., 2008). Some authors reported that high-substrate
consistency hydrolysis with supplementation of hemicellulose could be a practical solution for
minimizing end product-inhibition effects while producing a hydrolyzate with high glucose
concentration (Xiao et al., 2004).
1.3.3 Industrial significance
Cellulases and xylanases have major and numerous industrial applications, such as in pulp and paper
manufacturing, as well as in the textile industry for polishing of fabrics and laundry detergents for
improving fabric softness (Kuhad et al., 2011; Mansour et al., 2016). For example, Das et al. (2013)
used cellulases produced through an optimized solid-state fermentation for the deinking of waste pulp
of laser printed paper, that is, mainly the removal of chromophores and hydrophobic compounds. In
addition, cellulases are used in the extraction process of fruit and vegetable juices, starch processing
and formulations used for animal feeds (Dhillon et al., 2012a; Singhania et al., 2009). Cellulases also
have found promising applications for non-specific hydrolysis of chitosan to produce
chitooligosaccharides with low molecular weight, which showed high antibacterial activity (Xia et al.,
2008).
From biotechnological perspectives, the most important and recent application of cellulases is
bioenergy production. Cellulases are used to hydrolyze cellulosic wastes to sugars that can be
fermented to bioethanol or biogas, depending on the process conditions and microorganisms (Kuhad
et al., 2016). It is shown that there is a wide potential to develop a simple biological process to
produce bioethanol from a variety of lignocellulosic substrates, that is, by hydrolyzing and fermenting
carbohydrates, which are considered as waste materials produced in huge amounts especially in the
agro-industrial sector. The major constraint in the bioethanol process is precisely the costs of cellulase
production according to several studies (Arora et al., 2015; Lever, 2015). In order to reduce
downstream processing of the enzymatic compounds, some authors (Lever et al., 2010; Liu et al.,
2011; Rocha et al., 2013) directly applied the unprocessed cellulases obtained by SSF into hydrolytic
processes of organic wastes for further bioethanol production.
For biogas production and in the context of solid wastes, hydrolysis step is considered as rate limiting
and, therefore, it is considered as crucial for the efficiency: the more the substrate is hydrolyzed, the
greater is the amount of methane produced (Zverlov et al., 2015). The rate of decomposition during
the hydrolysis stage greatly depends on the amount and efficiency of hydrolytic enzymes produced by
the bacterial consortium. Moreover, the hydrolysis of cellulose (and hemicellulose) takes place slower
Chapter 1.Introduction
12
than other complex compounds, such as proteins (Kuhad et al., 2016), thus establishing the
importance of these enzymes in the process.
In both bioethanol and biogas production, process integration is perhaps one of the most important
approaches for the design of more intensive and cost-effective process configurations. Since most of
the current technologies only use a part of the organic wastes for obtaining bioproducts, using the
whole feedstock or optimally utilizing all the fractions present in the biomass will enhance the
economic and environmental feasibility of biofuels and co-products by a better understanding of
relationships between plant (bio)chemistry and potential bioproducts (Joyce & Stewart, 2012).
1.4 Solid-State Fermentation
In the context of circular economy, the most promising technology to produce low cost enzymatic
products using lignocellulosic biomass, is solid-state fermentation. Solid-state fermentation (SSF) has
been defined by many researchers. These definitions agree in the fact that this process involves solids
in the absence or near absence of free water (Chen, 2013; Thomas et al., 2013). In the SSF concept,
the solid substrate should possess enough moisture in absorbed or complex form to support microbial
growth and metabolic activities (Rabinovich et al., 2002; Singhania et al., 2010).
SSF is a three-phase system consisting of the continuous gas phase, the liquid film, and the solid
phase. It should be noted that there is no obvious relationship between the water content of the
substrate and the content of free water. Because of the strong hydraulic holding ability of some solid
substrates, such as that of the sugar beet plant material, even if the water content of the substrate is
more than 80%, there is seldom free water in the solid substrate. Consequently, the content of water
cannot be defined as the only standard of solid fermentation (Chen, 2013).
SSF has been classified in two categories depending on the type of the used microorganism: i) natural
SSF and ii) pure culture SSF; according to Krishna (2005). The conventional industrial processes
mostly involve pure cultures for the production of targeted products. On the other hand, mixed
cultures are preferred in the bioconversion processes of agro-industrial residues, such as composting
and ensiling.
In recent years, SSF has shown great potential in the development of several bioprocesses and
products. This technology has its own advantages over conventional submerged fermentation (SmF)
as presented in Table 1.1.
Chapter 1.Introduction
13
Table 1.1. Comparison between SSF and SmF. Source: El-Bakry et al., (2015).
Factor SSF SmF
Substrate Cheap and available e.g.
organic wastes
Expensive media ingredients
Inoculum Not necessary Essential
Aseptic conditions Not needed Essential
Moisture No free water Liquid media required
Agitation Very difficult Easy
Process control (T, pH) Difficult Easy
Enzyme yield High, due to high solid content Low, due to low solid content
Downstream processing Can be easy, cheap Difficult, expensive
Liquid waste Low quantities High quantities
Scale up Difficult, new designs required Easy, commercial equipments
available
Volume and costs of
equipment
Small reactors can be used,
low cost
Large scale reactors are required,
high cost
As stated, the potential of SSF is to provide the cultured microorganism an environment as close as
possible to natural environment where usually they exist and from where they are isolated. This is
apparently the main factor why microbes perform well and give higher products yields in SSF when
compared with SmF carried out at optimal conditions for microorganisms growth (Thomas et al.,
2013). As mentioned above, the use of agro-industrial residues and by-products as substrates in SSF
processes adds economic value to these materials and solves the problem of their disposal or
treatment. In this context, SSF has been useful for the production of several value added products,
using organic wastes as the substrate, as presented in Table 1.2.
As observed in Table 1.2, most of the research carried out in SSF is performed using few grams of
substrate. In addition, most of these studies have been performed under initial sterile conditions, using
specific microorganisms and systems with controlled temperature. Taking into account these
conditions, the scaling up of the process appears as an extremely difficult issue, considering the
widely known constraints of working with large amounts of solid substrates and the costs associated
to the conventional operational conditions. Most of these issues have been partially addressed by our
research group, working in a representative scale and taking into account the heterogeneous nature of
the solid organic wastes.
Chapter 1.Introduction
14
The composting research group, GICOM, has en extensive experience on the study of composting
processes. Using this "know-how" and in order to achieve a cost-effective and sustainable SSF
process, our research group has proposed to work in a self-heating process, under near adiabatic
conditions using either native or externally added microbiota to non sterile substrates.
Table 1.2.Products obtained using organic wastes as substrates by means of SSF.
Substrate Microorganism Product Amount (g) Reference
Sugarcane
bagasse
Kluyveromyces
marxianus
Inulinase 2,000 (Astolfi et al., 2011)
Winterization
oil cake
Native microbiota Lipases 2,500 (Santis-Navarro et al., 2011)
Cotton waste Aspergillus spp. Hydrolytic and
oxidative enzymes
5-20 (Bansal et al., 2012; Liu et al.,
2011)
Wheat waste Aspergillus spp. Cellulases, xylanases 5-40 (Bansal et al., 2012; Dhillon et
al., 2012b)
Hair waste Native microbiota Proteases 2,000 (Abraham et al., 2013)
Soy waste Native microbiota Proteases 2,000 (Abraham et al., 2013)
Orange peels Native microbiota Cellulases 15,000 (Maulini-Duran et al., 2015)
Empty palm
fruit bunches
and palm
kernel cake
Isolated oleaginous
fungi
Cellulase and lipids < 1 (Cheirsilp & Kitcha, 2015)
Apple
pomace
Yeast mixture Aromas 6,800 (Rodríguez Madrera et al.,
2015)
Plum pomace/
Distillery
wastes
Aspergillus níger /
Rhizopus
oligosporus
Antioxidants 15 (Dulf et al., 2016)
Aguamiel Aspergillus
oryzae
Fructooligosacharides/
Fructosyltransferase
n.r (Muñiz-Márquez et al., 2016)
Winterization
oil cake
Starmerella
bombicola
Sophorolipids 100 (Jiménez-Peñalver et al., 2016)
Chapter 1.Introduction
15
1.4.1 Process description
The proposal for a more sustainable SSF consisted of a process that follows the same pattern than the
composting process. This process is mainly comprised by two stages: high rate degradation stage and
curing, as presented in Figure 1.4.
a) High rate degradation stage: In general, the hydrolysis of complex organic matter into simple
organic and inorganic molecules takes place during this stage. This hydrolysis in carried out by
microorganisms at its highest metabolic activity, which is reflected in a high oxygen uptake rate.
Also, in this stage heat is produced as a result of the metabolic activity, since these are exothermic
biological processes (Haug, 1993). In the early stages of the process, mesophillic conditions are
present, therefore, mesophillic microorganisms thrive. These microorganisms use available oxygen to
oxidize carbon from the solid matrix into biomass, energy and other products, with the consequent
CO2 and water production. As the heat is generated due to metabolic activity, the temperature starts to
increase. When the system achieves 45ºC, the mesophilic microorganisms either die or remain with
basal metabolism until more suitable conditions are presented. At the same time, thermophilic
microorganisms are favoured, and quickly start to consume the readily available materials and to
colonize the system. Once this material is fully consumed, the biological activity of thermophilic
microorganisms decreases, which generates a decrease in the temperature, also named cooling stage
(Haug, 1993). This reduction provides the proper conditions for mesophilic microorganisms, which
start to re-colonize the reactor and continue the process.
The thermophilic stage is important for the composting process due to the sanitation that provides to
the solid matrix that will be fully composted. When considering a SSF productive process,
thermophilic conditions could led to a production of thermostable enzymatic compounds, as reported
by Santis-Navarro et al. (2011). However, temperature rising above 70ºC could derive in a severe
reduction of microorganisms populations, which may affect the productivity of the process. In
addition, at high temperatures an increase in ammonia emissions can occur, mainly generated from
stripping during ammonium degradation (Pagans et al., 2006). This is important from the
environmental point of view and further assessment of the gaseous stream treatment.
b) Curing: After the high rate degradation stage, the temperature reaches mesophilic levels, where
curing stage begins. During this stage, the oxygen requirements are notably lower than in the high rate
stage, which results in a net loss of organic and inorganic matter. The importance of this stage is the
formation of high molecular compounds resistant to biodegradation from dead biomass (Haug, 1993).
At the end of this stage, the solid material will be fully stabilized, with generation of CO2, water,
mineral ions and stabilized organic matter, mostly in the form of humic acids.
Chapter 1.Introduction
16
Figure 1.4. Schematic temperature profile and main characteristics of a classic composting process.
Two processes for the production of hydrolytic enzymes have been studied by means of a self-heating
process working under near adiabatic conditions. In this sense, lipases and proteases production took
place during the high rate degradation stage (Abraham et al., 2013; Santis-Navarro et al., 2011). This
may imply that for hydrolytic enzymes production, the curing stage is probably not relevant. The SSF
should be then stopped in the high rate stage to extract the maximum enzymatic activity. Fermented
solids could be further stabilized to a compost-like product.
1.4.2 Environmental aspects
As any other productive process, SSF development should take into consideration environmental
aspects. The most representative aspects are disposal or treatment of the produced solid wastes and
released the gaseous emissions.
When a targeted compound is obtained by means of SSF, it is often extracted from the solid matrix.
This extraction generates a waste stream of fermented spent solid, that will require stabilization for a
potential utilization as new substrate (Lever, 2015; Thomas et al., 2013). This new substrate can be
Chapter 1.Introduction
17
treated in order to obtain an organic soil amendment or even bioenergy, such as biogas or bioethanol.
These processes would close the cycle of organic matter, in accordance to the circular economy
paradigm (Lever, 2015). In this context, Abraham et al. (2013) worked with a SSF process for
proteases production. Proteases were obtained after extraction from the solid matrix, generating a
fermented spent solid as a waste stream. This waste was post-treated, achieving full stabilization after
14 days of composting. Also, Maulini-Duran et al. (2015) assessed the stabilization of fermented
spent solids from SSF for proteases, cellulases and lipases, achieving similar results.
A step forward on the complete reuse of wastes with environmental and economic assessment was
taken by Lever (2015). In this study, a full ethanol plant was designed using the "zero waste"
approach as shown in Figure 1.5. This author compared the use of commercial cellulases and
cellulases produced by SSF using wheat straw and fermented solid from ethanol fermentation. The
produced fermented solids from SSF were stabilized by anaerobic digestion with the consequent
biogas production and further composting process, with full solid recycling. This study showed that
the proposed process resulted in high energy yield ratios, a net surplus of on-site heat and electricity,
and substantial reductions in the energy required to produce and transport the cellulase compared to
commercial preparations. Recycling fermented solids from ethanol fermentation as substrate for
cellulase production resulted in a reduction in energy consumption and a decrease in energy yield
ratio compared to using fresh wheat straw.
Figure 1.5. Unit operations for proposed cellulose to ethanol conversion process. Source: (Lever,
2015).
One other aspect to consider is the gaseous emissions. Gaseous emissions play an important role in
the environmental impact of processes used to manage solid organic substrates. Greenhouse gases
Chapter 1.Introduction
18
(GHG), ammonia (NH3) and volatile organic compounds (VOC) are the main gases emitted during
SSF processes (Maulini-Duran et al., 2015). Also, CH4 and N2O are often released, which present a
global warming potential (GWP) of 34 and 298 kg CO2 equivalents, respectively (IPCC, 2013).
Ammonia is not considered to be a direct GHG, but contributes to global warming because once
deposited in the soil, it can be converted into N2O after nitrification and denitrification reactions
(Wunderlin et al., 2013). Due to its high GWP, N2O can contribute strongly to the carbon footprint,
thus it is important to understand its formation during the SSF.
In spite of the great importance of this subject, there are only two reported works on the gaseous
emission for SSF process (Gassara et al., 2011; Maulini-Duran et al., 2015). While Maulini-Duran et
al., (2015) assessed gaseous emissions of three different SSF processes, Gassara et al. (2011) explored
the effect of different waste management systems of apple pomace and the gaseous emissions
produced. The different scenarios evaluated were incineration, landfill; animal feed; enzyme
production by SSF and composting. After a life cycle assessment study, these authors showed that
enzyme production (906.81 tons of CO2 equivalent per year) and animal feed (963 tons of CO2
equivalent per year) were the least polluting options in terms of GHG emissions followed by
incineration (1122 tons of CO2 equivalent per year), composting (1273tons of CO2 equivalent per
year) and landfill (1841 tons of CO2equivalent per year). These results confirm another advantage of
SSF, not only as an effective productive process, but as an environmentally friendly process. For this
reason, more studies on gaseous emissions are required.
1.5 Cellulase production by SSF
As mentioned in the previous sections, cellulases are a group of enzymes with high biotechnological
interest. The possibility to obtain this enzyme by means of SSF using organic solid wastes is very
interesting due to the economical and environmental advantages.
Cellulase production is generally represented as three enzymatic activities of filter paper (FPase) for
cellobiohydrolase or exoglucanase, carboxymethylcellulase (CMCase) for endoglucanase and β-
glucosidase (BGase) for cellobiase or β -glucosidase. These enzymatic activities are normally
expressed as units of activity per grams of dry matter, i.e. FPU g-1DM for FPase and U g-1DM for
CMCase and BGase respectively, when produced by SSF.
Chapter 1.Introduction
19
1.5.1 Process conditions
The most utilized substrates for cellulase production are agricultural wastes (soybean, wheat, rice,
corn, cotton, sugarcane bagasse), fruits (apple, orange and banana peels), as well as residues from
wood industries (softwood or hardwood) as shown in Table 3.
In most of the work developed on SSF using organic solid wastes, the addition of nutrients is often
required in order to stimulate the growth of a single or mixed microorganisms. However, in some
cases, depending on the nature of the substrate, there is no need of exogenous nutrients addition, as
reported by Dhillon et al. (2011) when using wheat bran as substrate. In this sense, many researchers
have worked on evaluating mixtures of substrates aiming to either increase cellulase production or, at
least, achieve a reduction of additional nutrient supplementation (Brijwani et al., 2010; El-Bakry et
al., 2015; Soni et al., 2010).
In addition to the substrate, it is crucial, as in any biological process, to select the proper
microorganism(s). In this context, bacteria and fungi are able to use cellulose as primary carbon
source; however, bacteria are incapable to fully degrade crystalline cellulose since their cellulase
system is incomplete. For fungi the situation is different, these microorganisms are able to completely
degrade this structure. It is for this reason that many fungi are studied in recent research works,
especially from the genera Trichoderma and Aspergillus. Some species of these two fungal genera
have been reported as the most important cellulase producer microorganisms applied to SSF processes
(Farinas, 2015; Hansen et al., 2015).
It is important to mention that some fungi can produce preferently one enzyme over the others. For
instance, Thrichoderma reesei is well known for its potential for the hyper-production of cellulase,
but usually has low activity or is deficient in the production of β-glucosidase (Brijwani et al., 2010;
Brijwani et al., 2011). The β-glucosidase hydrolyzes cellobiose, which otherwise will inhibit cellulase
joint action and will result in overall low substrate hydrolysis rate. It is for this reason that some
authors have made great efforts to improve inoculum and substrates mixtures to increase β-
glucosidase production (Dhillon et al., 2012a; Dhillon et al., 2011; Hu et al., 2011; Kaur et al., 2012;
Liu et al., 2011).
In terms of operational parameters for cellulase production, there are critical aspects to take into
account: availability of oxygen, temperature, pH, water activity and moisture, bed properties and
nature of the substrate have been reported as the most important parameters (Mekala et al., 2008;
Olofsson et al., 2008; Van Dyk & Pletschke, 2012). In this sense, many authors have targeted their
research on the optimization of these parameters by means of response surface methodology and
Chapter 1.Introduction
20
mathematical modelling techniques (Brijwani et al., 2011; Das et al., 2013; Dhillon et al., 2012a;
Khanahmadi et al., 2006).
It has been reported that optimal pH range for cellulase production is between 4-8, depending on
which cellulose activity is measured and the microorganism used (Bahrin et al., 2012; Lee et al.,
2011; Mekala et al., 2008; Soni et al., 2010; Yang et al., 2012). In general, pH has a soft effect on
exoglucanase and β -glucosidase activity within most of experimental research. Robustness against
changes of pH from its initial value during enzyme production would be benefitial in shielding any
effect on enzyme activity due to pH fluctuations during the productive process. For instance, Brijwani
et al. (2010) operated SSF for cellulase production using wheat bran as substrate and using
Trichoderma reseei as inoculum. This fermentation started at optimum conditions for the inoculum,
however the process was carried out without pH control. The microorganism was able not only to
survive, but to sustain growth and enzymatic production even working at high alkaline conditions.
Also, many authors have stated that cellulase stability at different pH is wide and it has been reported
in a ratio between 4-8, maintaining a residual activity in the range of 60-100% (Bansal et al., 2012;
Hu et al., 2011; Liu et al., 2011). Above the mentioned pH, the range of cellulase activity decreases
drastically.
The process temperature used during the process is an important factor affecting microbial growth and
activity and thus, enzyme production in SSF. Commonly, temperature ranges between 25-35°C (Das
et al., 2013); however, cellulolytic activity has been observed in a wide range of values, as presented
in Table 3. Most of the maximum cellulase production has been found at 30°C, but appreciable
production has been obtained in thermophillic conditions. Temperatures in the range of 40-50°C have
reported to produce thermostable cellulases with enzymatic activities ranging between 224-526 U g-
1DM for CMCase, 7.2-144.6 FPU g-1DM for FPase and 22.4 -30 U g-1DM for β-glucosidase (Bansal
et al., 2012; Liu et al., 2011).
Another important aspect to consider is the moisture content during SSF. It has been well established
that both high and low moisture content can directly affect the productivity of the SSF. High moisture
content decreases porosity, promoting compaction and lowering oxygen transfer. On the other hand,
lower moisture content causes a reduction on the solubility of nutrients of the solid substrate, a lower
degree of sweeling and higher water tension (Anto et al., 2006). Optimal moisture content ranges
between 40-60% (Bansal et al., 2012; Chandra et al., 2007; Das et al., 2013; Dhillon et al., 2012b;
Liu, 2007; Soni et al., 2010; Sukumaran et al., 2009) using different lignocellulosic substrates.
Regardless, some authors have obtained higher levels at 70%-80% moisture content reaching FPase
activity values between 2.5-5.0 FPU g-1DM (Lee et al., 2011; Liu et al., 2011).
Chapter 1.Introduction
21
As discussed in section 1.3.2, there are many factors affecting cellulose hydrolysis that will require
assessment in order to achieve proper cellulase production. Some authors have worked in different
strategies to achieve that goal, such as alkaline or acid pretreatment or inductors addition (Bansal et
al., 2012; Dhillon et al., 2012a; Sanghvi et al., 2010; Yang et al., 2012). Regarding the pretreatments,
Bansal et al. (2012) determined that cellulase production using different raw materials was lower than
when using the pretreated substrates. This fact was attributed to the higher lignin content and the firm
binding, making cellulose less accessible to microorganisms. Also, the highest cellulase production
has been found when using alkaline pretreatment. Acid pretreatment solubilized hemicellulose
structure, releasing monomers, furfural and other volatile compounds which, along with lignin,
produce a negative environment for an organism to grow, thereby lowering productivities and yields
of the process. In contrast, Sanghvi et al. (2010), using the same pretreatment, showed no
improvement on cellulase production, which was attributed to a decrease in mycelia growth on the
media, which led to a rapid decrease on enzyme production. In this sense, the hemicellulase
production is often study along with cellulase production, in order to obtain a mixed enzymatic pool
that can potentially improve hydrolysis in a bioethanol production process. Table 3 shows several
cellulase producing processes and most of them included hemicellulase production represented as
xylanase.
Higher cellulase activity values were obtained by adding inductors (Dhillon et al., 2012a), reaching
values of 224-526.3 U g-1DM for CMCase, 30.4-144.6 FPU g-1DM for FPase, 99 U g-1DM for β -
glucosidase and 3106 U g-1DM of xylanase. However, the feasibility to implement the addition of
inductors, depends on economical aspects such as availability and cost of the selected inductor.
All the above mentioned researches have been made on laboratory scale in a single batch
configuration, using amounts of substrate between 5 to 400 g (Table 1.3). Using low amounts of
heterogeneous material provides a non easily scalable system as it can avoid compaction problems
and its consequent poor oxygen transfer, which are very important factors to take into account when
scaling up and reproducibility issues are assessed.
Chapter 1.Introduction
22
Table 1.3. Summary of latest cellulase production by solid state fermentation using wastes as
substrate. RS: rice straw, RH: rice husk, WB: wheat bran, SH: soybean hull, SCB: sugar cane bagasse,
AP: apple pomace, BSG: brewers spent grain., T: temperature, M: moisture, FPase: filter paper,
CMCase: endocellulase, BGase: β-glucosidase and Xyl: xylanase.
Inoculum Substrate Process conditions Cellulase
activity
Reference
Type Type
Scale (SSF time)
pH T (°C)
M (%)
Max. Activity (U g-1DM)
T. reesei
A.oryzae
SH Deep bed
reactor
5.0 30 70 FPase 5.4
CMCase 58.6
BGase 18.4
Xyl 242
(Brijwani et al., 2011)
T. reesei
A.oryzae
SH:WB Static tray
reactor
(96h)
5.0 30 70 FPase 10.8
CMCase 100.7
BGase 10.7
(Brijwani et al., 2010)
T. reseei
A. niger
RS:WB Tray
(120h)
- 30 - FPase 35.8
CMCase 132.3
BGase 33.7
Xyl 3106
(Dhillon et al., 2011)
A. niger AP:RH
(48h)
Lab - 30 75 FPase 133
CMCase 172
(Dhillon et al., 2012b)
T. reesei RS: SCB
(72h)
Lab - 30 40-
57
FPase 4.6
CMCase 135.4
BGase 21.5
(Sukumaran et al.,
2009)
Trichoderma
sp.
GIM 3.0010
AP Lab
(120h)
- 30 70 FPase 1-7.6 (Sun et al., 2010)
Trichoderma
veride
sp.3.2942
RH:WB
Lab
(120h)
- 30 - FPase 6.3 (Hu et al., 2011)
Chapter 1.Introduction
23
Table 1.3. (Continued). RS: rice straw, RH: rice husk, WB: wheat bran, SH: soybean hull, SCB:
sugar cane bagasse, AP: apple pomace, BSG: brewers spent grain., T: temperature, M: moisture,
FPase: filter paper, CMCase: endocellulase, BGase: β-glucosidase and Xyl: xylanase.
Inoculum Substrate Process conditions Cellulase
activity
Reference
Type Type
Scale (SSF time)
pH T (°C)
M (%)
Max Activity (U g-1DM)
A.niger
NRRL 567
AP:BSG
Lab
(96h)
- 30 - FPase 71.1
CMCase 103.9
BGase 55.9
Xyl 1618.1
(Dhillon et al., 2012a)
A.niger
NRRL 567
AP:BSG
Lactoserum
Lab
(96h)
- 30 - FPase 96.7
CMCase 134.4
BGase 90.1
Xyl 2619.5
(Dhillon et al., 2012a)
A. niger
NRRL 2001
AP:BSG
Lab
(96h)
- 30 - FPase 85.6
CMCase 107.7
BGase 32.8
Xyl 1110.9
(Dhillon et al., 2012a)
A. niger
NRRL 2001
AP:BSG
Lactoserum
Lab
(96h)
- 30 - FPase 130.4
CMCase 148.9
BGase 90.1
Xyl 2281.7
(Dhillon et al., 2012a)
A. niger
NS-2
Corn cob
Lab
(96h)
6.5 30 60 FPase 3.1
CMCase 10
BGase 1.8
(Bansal et al., 2012)
A. niger
NS-2
Carrot
peelings
Lab
(96h)
6.5 30 60 FPase 0.3
CMCase 4.9
BGase 5.0
(Bansal et al., 2012)
Chapter 1.Introduction
24
Table 1.3. (Continued). RS: rice straw, RH: rice husk, WB: wheat bran, SH: soybean hull, SCB:
sugar cane bagasse, AP: apple pomace, BSG: brewers spent grain., T: temperature, M: moisture,
FPase: filter paper, CMCase: endocellulase, BGase: β-glucosidase and Xyl: xylanase.
Inoculum Substrate Process conditions Cellulase
activity
Reference
Type Type
Scale (SSF time)
pH T (°C)
M (%)
Max Activity (U g-1DM)
A. niger NS-
2
Kitchen
waste
Lab
(96h)
6.5 30 60 FPase 10.3
CMCase 48.6
BGase 19.5
(Bansal et al., 2012)
A. niger NS-
2
RH
Lab
(96h)
6.5 30 60 FPase 3.1
CMCase 14.1
BGase 10.6
(Bansal et al., 2012)
A. fumigatus Vinegar
industry
waste
Lab 4.0 50 80 FPase 144.6
CMCase 526.3
(Liu et al., 2011)
A. fumigatus RS
Lab(
5d)
7.0 45 75 FPase 3.4
CMCase 98.5
BGase 250.9
Xyl 2782
(Soni et al., 2010)
A. fumigatus WB Lab 7.0 45 75 FPase 0.67
CMCase 20
BGase 99.4
Xyl 1722
(Soni et al., 2010)
T.reesei
CBS 439.92
WS Lab - 30 - FPase 0.6 (Lever et al., 2010)
A. fumigatus RS:WB
(5d)
Lab 7.0 45 75 FPase 2.1
CMCase 40.9
BGase 243.7
Xyl 444
(Soni et al., 2010)
Chapter 1.Introduction
25
In summary, it is remarkable to mention that most of the research performed on cellulase production
by SSF has been carried out using pure strains and working under mesophillic conditions. In addition,
as observed in Table 3, the most common configuration is batch and using small amounts of substrate.
In this sense, there is a great challenge in optimize all of these aspects in order to develop a more
sustainable SSF process.
1.5.2 Novel microbiological approaches: Strain development and Metagenomics
Cellulases are expressed by a wide spectrum of microorganisms in nature. Screening and isolation of
cellulase-producing microorganisms from different environments is one of the most important ways to
produce novel cellulases. These newly screened microbes are sources of new cellulase genes with
diverse properties (Juturu & Wu, 2014).
In this context, Jing et al. (2015) searched for efficient fungal producers of cellulases in a natural
decay system, using a screening process containing hydrolyzed sugarcane bagasse as carbon source.
They successfully isolated 12 fungal species, which showed considerable amounts of exoglucanase
activity. These species were identified as Penicillium oxalicum strains and were assessed as inoculum
for pretreated sugarcane bagasse for further bioethanol production with promising results. Another
successful report was the presented by Ang et al., (2013). These authors were able to isolate a fungal
strain form cow dung. This strain was identified as Aspergillus fumigatus SK1 and was evaluated as
inoculum in a SSF process using oil palm trunk as substrate. Cellulase activities obtained were 54.3,
3.4, 4.5 and 418.7 U g-1DM substrates for endoglucanase (CMCase), exoglucanase (FPase), β -
glucosidase and xylanase respectively. These results are in the middle range of the reported in the
literature, as seen in Table 3. Many similar approaches are reported by Juturu & Wu (2014);
interestingly, most of them aimed to optimize cellulose degradation for bioethanol production.
Although the above mentioned approaches presented good results, the extreme complexity of natural
environments makes culture-based approaches extremely time and labor-consuming. Furthermore, the
results are based on phenotypic characteristics, which can be imprecise and/or inconclusive (Wang et
al., 2009). In this sense, many authors are currently studying novel techniques, such as strain
development, heterologous cellulase expression or metagenomics (Ray & Behera, 2017). The most
promising tool for non-sterile SSF application is metagenomics.
Metagenomics, the study of genetic material recovered directly from environmental samples, has the
potential to exploit the inexhaustible source of novel biocatalysts trapped in genomes of unculturable
Chapter 1.Introduction
26
microorganisms. This technology is a useful technique for the monitoring of microbial communities
in a SSF process or to discover enzymes from highly diverse environments (Li et al., 2016a).
To the best of our knowledge, there are only two research papers published using the metagenomics
approach to follow a multispecie SSF (Li et al., 2016b; Wang et al., 2016). Both of these reports were
focused on the study of patterns/dynamics of the populations and their functionality in the productive
process. Li et al. (2016) reported the effect of temperature of SSF for cereal starters production at
industrial scale. Using this technique, the authors obtained a deep understanding of the process that
allowed the effective control of the fermentative process by adjusting relevant environmental
parameters. On the other hand, Wang et al. (2016) assessed the influence of the different
microorganisms on the aroma and flavour production, during the SSF of a mixed substrate (wheat
bran, mash alcohol and chaff) for chinese vinegar production. The authors reported that an
improvement of the quality of the product can be obtained by working with the dataset generated by
the metagenomics of the process.
GICOM proposal of SSF includes a microbial succession of microorganism characteristic of the
composting process. There are a few reports of the follow up of microbial communities during the
composting of lignocellulosic materials (Langarica-Fuentes et al., 2014; López-González et al.,
2015a; López-González et al., 2015b).
López-González et al. (2015a) reported the microbial population succession and the enzymatic
activities expressed throughout the composting process of lignocellulosic material. In this work, more
than 4000 strains were isolated, enzymatically characterized, and identified by partial sequencing of
16S rRNA. Mesophilic bacteria were found during the entire process, while thermophilic conditions
gathered the highest total counts and spore-forming bacteria prevailed at the high rate degradation
stage. The authors also found that cellulolytic and ligninolytic bacteria appeared during thermophilic
stage but also appeared in considerable amounts at the end of the process. Furthermore, López-
González et al. (2015b) found that cellulolytic activity (obtained from cultured isolates) expressed by
the fungal community was also observed during thermophilic stage. In this sense, stable compost
could be a potential source of cellulase degraders for further isolation or direct application.
Although these reports are very promising, they are carried out using cultured microorganism, which
leave the uncultured microorganism without analysis. Due to the potential of utilization as source of
cellulolytic microorganism, it is interesting to assess the changes in uncultured microorganisms
population using a metagenomics approach.
Chapter 1.Introduction
27
1.5.3 Challenges
Once the proper inoculum, substrate and process conditions are selected and optimized, one of the
major challenges of developing a reproducible SSF processes is to work at a larger scale.
Most of the research performed has been carried out at laboratory scale (Table 3), usually working in
flasks for microorganism growth, which can be convenient for substrate/inoculum screening but
provides no useful information for posterior scaling up. In Table 1.4, a comparative of experiments
performed in lab and large scale is presented. In this sense, besides oxygen transfer, factors that need
to be considered include temperature, water content, morphology of microorganisms (especially in the
case of fungi) and sterilization requirements (Durand, 2003). Several reactor designs have been
proposed for cellulolytic enzyme production, including tray-type, packed-bed, and horizontal rotating
drum bioreactors, amongst others (Mitchell et al., 2006; Thomas et al., 2013). Each one of these
designs has its own advantages and disadvantages, which suggested the necessity to develop novel
bioreactors with better design in order to solve the major problems related to the scale-up processes
for the production of enzymes through SSF.
There are few recent studies on the overcoming of some of the disadvantages and limitations
regarding the SSF scaling-up processes. An extensive analysis on the design and operation of
bioreactors in SSF has been published by Thomas et al. (2013). Cellulase production using these
designs has been successfully carried out by several researchers (Brijwani et al., 2011; Dhillon et al.,
2011; Hansen et al., 2015). Even these approaches are a good alternative, there is still a requirement
for temperature control and heat removal. One different approach has been reported by GICOM
research group for protease and lipase (Abraham et al., 2013; Santis-Navarro et al., 2011). In these
approaches, the temperature was not controlled and enzymes were produced in a batch fermentation
process similar to that of composting. The temperature rose to thermophilic values due to heat
released and decreased to ambient values during the fermentation of the readily biodegradable matter.
Another challenge to overcome is the possibility to develop a process in a continuous configuration.
Although Abraham et al. (2013) and Santis-Navarro et al. (2011) approaches were successful and
assessed in a pilot scale (Maulini-Duran et al., 2015), they were developed in a batch configuration
which could prevent the process of being economically feasible. Very few attempts have been
reported on continuous SSF. Two reports have been published on enzyme production in batch and
fed-batch configuration for inulinase (Astolfi et al., 2011) and cellulase (Cheirsilp & Kitcha, 2015).
Both researches proved that enzymatic production was sustained in time and, in some cases,
enhanced. These results were very promising for enzyme production and lignocellulosic waste
management, however only the inulinase experiment was performed at a representative scale (2 Kg).
Chapter 1.Introduction
28
Cellulase experiments were carried out using less than 1 g of material, which is not representative for
further scale up. In this context, it is necessary to undertake more studies to develop a continuous
system for enzymatic production.
Table 1.4. Conditions needed for lab scale vs. large or commercial scale SSF. Source: El-Bakry et al.
(2015).
Condition Lab scale Large scale
pH control
T control, Heat removal
Possible by pH adjustment
Easy, e.g. water bath
Not possible
Possible by aeration, costly,
presence of T gradients.
Solid handling Very easy Very difficult
Inoculation Easy, not expensive High costs, difficult
homogenization
Agitation Very easy Possible but costly, e.g rotatory
drums
Aeration Sufficient Moderate-high, expensive
References
Abraham, J., Gea, T., Sánchez, A. 2013. Potential of the solid-state fermentation of soy fibre residues by native microbial populations for bench-scale alkaline protease production. Biochemical Engineering Journal, 74, 15-19.
Agbor, V., Cicek, N., Sparling, R., Berlin, A., Levin, D. 2011. Biomass pretreatment: Fundamentals toward application. Biotechnology Advances, 29(6), 675-685.
Allesch, A., Brunner, P. 2014. Assessment methods for solid waste management: A literature review. Waste Management Research, 32(6), 461-73.
Ang, S., Shaza, E., Adibah, Y., Suraini, A., Madihah, M. 2013. Production of cellulases and xylanase by Aspergillus fumigatus SK1 using untreated oil palm trunk through solid state fermentation. Process Biochemistry, 48(9), 1293-1302.
Anto, H., Trivedi, U., Patel, K. 2006. Glucoamylase production by solid-state fermentation using rice flake manufacturing waste products as substrate. Bioresource Technology, 97(10), 1161-1166.
Arora, R., Behera, S., Kumar, S. 2015. Bioprospecting thermophilic/thermotolerant microbes for production of lignocellulosic ethanol: A future perspective. Renewable and Sustainable Energy Reviews, 51, 699-717.
Astolfi, V., Joris, J., Verlindo, R., Oliveira, J., Maugeri, F., Mazutti, M., de Oliveira, D., Treichel, H. 2011. Operation of a fixed-bed bioreactor in batch and fed-batch modes for production of inulinase by solid-state fermentation. Biochemical Engineering Journal, 58–59, 39-49.
Bahrin, E., Ibrahim, M., Abd Razak, M., Abd-Aziz, S., Shah, U., Alitheen, N., Salleh, M. 2012. Improved cellulase production by Botryosphaeria rhodina from OPFB at low level moisture
Chapter 1.Introduction
29
condition through statistical optimization. Preparative Biochemistry and Biotechnology, 42(2), 155-170.
Bansal, N., Tewari, R., Soni, R., Soni, S. 2012. Production of cellulases from Aspergillus niger NS-2 in solid state fermentation on agricultural and kitchen waste residues. Waste Management, 32(7), 1341-1346.
Berghem, L., Pettersson, L. 1973. The mechanism of enzymatic cellulose degradation. Purification of a cellulolytic enzyme from Trichoderma viride active on highly ordered cellulose. European Journal of Biochemistry, 37(1), 21-30.
Bidlack, J., Malone, M., Benson, R. 1992. Molecular structure and component integration of secondary cell walls in plants. Proceedings of the Oklahoma Academy of Science., 72, 51-56.
Brethauer, S., Wyman, C. 2010. Review: Continuous hydrolysis and fermentation for cellulosic ethanol production. Bioresource Technology, 101(13), 4862-4874.
Brijwani, K., Oberoi, H., Vadlani, P. 2010. Production of a cellulolytic enzyme system in mixed-culture solid-state fermentation of soybean hulls supplemented with wheat bran. Process Biochemistry, 45(1), 120-128.
Brijwani, K., Vadlani, P., Hohn, K., Maier, D. 2011. Experimental and theoretical analysis of a novel deep-bed solid-state bioreactor for cellulolytic enzymes production. Biochemical Engineering Journal, 58–59, 110-123.
Chandra, M., Viswanath, B., Reddy, B. 2007. Cellulolytic enzymes on lignocellulosic substrates in solid state fermentation by Aspergillus niger. Indian Journal of Microbiology, 47(4), 323-328.
Cheirsilp, B., Kitcha, S. 2015. Solid state fermentation by cellulolytic oleaginous fungi for direct conversion of lignocellulosic biomass into lipids: Fed-batch and repeated-batch fermentations. Industrial Crops and Products, 66, 73-80.
Chen, H. 2013. Modern Solid State Fermentation: Theory and Practice. 1 ed, (Ed.) H. Chen, Springer. New York.
Das, A., Paul, T., Halder, S., Jana, A., Maity, C., Das, P., Pati, B., Mondal, K. 2013. Production of cellulolytic enzymes by Aspergillus fumigatus ABK9 in wheat bran-rice straw mixed substrate and use of cocktail enzymes for deinking of waste office paper pulp. Bioresource Technology, 128, 290-296.
Davies, G., Henrissat, B. 1995. Structures and mechanisms of glycosyl hydrolases. Structure, 3(9), 853-859.
Dhillon, G., Brar, S., Kaur, S., Metahni, S., M'Hamdi, N. 2012a. Lactoserum as a moistening medium and crude inducer for fungal cellulase and hemicellulase induction through solid-state fermentation of apple pomace. Biomass and Bioenergy, 41, 165-174.
Dhillon, G., Kaur, S., Brar, S., Verma, M. 2012b. Potential of apple pomace as a solid substrate for fungal cellulase and hemicellulase bioproduction through solid-state fermentation. Industrial Crops and Products, 38(1), 6-13.
Dhillon, G.S., Oberoi, H.S., Kaur, S., Bansal, S., Brar, S.K. 2011. Value-addition of agricultural wastes for augmented cellulase and xylanase production through solid-state tray fermentation employing mixed-culture of fungi. Industrial Crops and Products, 34(1), 1160-1167.
Dulf, F., Vodnar, D., Socaciu, C. 2016. Effects of solid-state fermentation with two filamentous fungi on the total phenolic contents, flavonoids, antioxidant activities and lipid fractions of plum fruit (Prunus domestica L.) by-products. Food Chemistry, 209, 27-36.
Durand, A. 2003. Bioreactor designs for solid state fermentation. Biochemical Engineering Journal, 13(2–3), 113-125.
El-Bakry, M., Abraham, J., Cerda, A., Barrena, R., Ponsá, S., Gea, T., Sánchez, A. 2015. From Wastes to High Value Added Products: Novel Aspects of SSF in the Production of Enzymes. Critical Reviews in Environmental Science and Technology, 45(18), 1999-2042.
Farinas, C. 2015. Developments in solid-state fermentation for the production of biomass-degrading enzymes for the bioenergy sector. Renewable and Sustainable Energy Reviews, 52, 179-188.
Gassara, F., Brar, S., Pelletier, F., Verma, M., Godbout, S., Tyagi, R. 2011. Pomace waste management scenarios in Québec—Impact on greenhouse gas emissions. Journal of Hazardous Materials, 192(3), 1178-1185.
Chapter 1.Introduction
30
Gusakov, A., Sinitsyn, A. 1992. A theoretical analysis of cellulase product inhibition: Effect of cellulase binding constant, enzyme/substrate ratio, and β-glucosidase activity on the inhibition pattern. Biotechnology and Bioengineering, 40(6), 663-671.
Hansen, G., Lübeck, M., Frisvad, J., Lübeck, P., Andersen, B. 2015. Production of cellulolytic enzymes from ascomycetes: Comparison of solid state and submerged fermentation. Process Biochemistry, 50(9), 1327-1341.
Haug, R. 1993. The Practical Handbook of Compost Engineering, Lewis Publishers. Boca Raton, Florida.
Hu, T., Zhou, Y., Dai, L., Wang, Y., Liu, D., Zhang, J., Liu, H. 2011. Enhanced cellulase production by solid state fermentation with polyurethane foam as inert supports. Procedia Engineering, 18(0), 335-340.
IPCC. 2013. Contribution of Working Group Anthropogenic and Natural Radiative Forcing to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge University Press.
Jacquet, N., Haubruge, E., Richel, A. 2015. Production of biofuels and biomolecules in the framework of circular economy: A regional case study. Waste Management Research, 33(12), 1121-6.
Jiménez-Peñalver, P., Gea, T., Sánchez, A., Font, X. 2016. Production of sophorolipids from winterization oil cake by solid-state fermentation: Optimization, monitoring and effect of mixing. Biochemical Engineering Journal, 115, 93-100.
Jing, L., Zhao, S., Xue, J., Zhang, Z., Yang, Q., Xian, L., Feng, J. 2015. Isolation and characterization of a novel Penicillium oxalicum strain Z1-3 with enhanced cellobiohydrolase production using cellulase-hydrolyzed sugarcane bagasse as carbon source. Industrial Crops and Products, 77, 666-675.
Joyce, B., Stewart, C. 2012. Designing the perfect plant feedstock for biofuel production: using the whole buffalo to diversify fuels and products. Biotechnology Advances, 30(5), 1011-22.
Juturu, V., Wu, J. 2014. Microbial cellulases: Engineering, production and applications. Renewable and Sustainable Energy Reviews, 33, 188-203.
Kaur, S., Dhillon, G., Brar, S., Chauhan, V. 2012. Carbohydrate degrading enzyme production by plant pathogenic mycelia and microsclerotia isolates of Macrophomina phaseolina through koji fermentation. Industrial Crops and Products, 36(1), 140-148.
Khanahmadi, M., Mitchell, D., Beheshti, M., Roostaazad, R., Sánchez, L. 2006. Continuous solid-state fermentation as affected by substrate flow pattern. Chemical Engineering Science, 61(8), 2675-2687.
Kostylev, M., Wilson, D. 2012. Synergistic interactions in cellulose hydrolysis. Biofuels, 3(1), 61-70. Krishna, C. 2005. Solid-state fermentation systems-an overview. Critical Reviews of Biotechnology,
25(1-2), 1-30. Kuhad, R., Deswal, D., Sharma, S., Bhattacharya, A., Jain, K., Kaur, A., Pletschke, B., Singh, A.,
Karp, M. 2016. Revisiting cellulase production and redefining current strategies based on major challenges. Renewable and Sustainable Energy Reviews, 55, 249-272.
Kuhad, R., Gupta, R., Singh, A. 2011. Microbial cellulases and their industrial applications. Enzyme Res, 2011, 280696.
Langarica-Fuentes, A., Zafar, U., Heyworth, A., Brown, T., Fox, G., Robson, G. 2014. Fungal succession in an in-vessel composting system characterized using 454 pyrosequencing. FEMS Microbiology Ecology, 88(2), 296-308.
Lee, C., Darah, I., C., I. 2011. Production and Optimization of Cellulase Enzyme Using Aspergillus niger USM AI 1 and Comparison with Trichoderma reesei via Solid State Fermentation System. Biotechnology Research International, 2011, 1-6.
Lee, S., Shin, H., Ryu, D., Mandels, M. 1982. Adsorption of cellulase on cellulose: Effect of physicochemical properties of cellulose on adsorption and rate of hydrolysis. Biotechnology and Bioengineering, 24(10), 2137-53.
Leite, R., Alves-Prado, H., Cabral, H., Pagnocca, F., Gomes, E., Da-Silva, R. 2008. Production and characteristics comparison of crude β -glucosidases produced by microorganisms
Chapter 1.Introduction
31
Thermoascus aurantiacus and Aureobasidium pullulans in agricultural wastes. Enzyme and Microbial Technology, 43(6), 391-395.
Lever, M. 2015. Modelling the energy performance of a farm-scale cellulose to ethanol process with on-site cellulase production and anaerobic digestion. Renewable Energy, 74, 893-902.
Lever, M., Ho, G., Cord-Ruwisch, R. 2010. Ethanol from lignocellulose using crude unprocessed cellulase from solid-state fermentation. Bioresource Technology, 101(18), 7083-7087.
Li, B., Zhu, Y., Gu, Z., Chen, Y., Leng, J., Gou, X., Feng, L., Li, Q., Xi, D., Mao, H., Yang, S. 2016a. Screening and characterization of a novel ruminal cellulase gene (Umcel-1) from a metagenomic library of gayal (Bos frontalis). Journal of Integrative Agriculture, 15(4), 855-861.
Li, M., Fan, Y., Xu, F., Sun, R., Zhang, X. 2010. Cold sodium hydroxide/urea based pretreatment of bamboo for bioethanol production: Characterization of the cellulose rich fraction. Industrial Crops and Products, 32(3), 551-559.
Li, P., Lin, W., Liu, X., Wang, X., Luo, L. 2016b. Environmental Factors Affecting Microbiota Dynamics during Traditional Solid-state Fermentation of Chinese Daqu Starter. Frontiers in Microbiology, 7, 1237.
Li, Y., Sun, Z., Ge, X., Zhang, J. 2016c. Effects of lignin and surfactant on adsorption and hydrolysis of cellulases on cellulose. Biotechnology for Biofuels, 9(1), 20.
Liguori, R., Amore, A., Faraco, V. 2013. Waste valorization by biotechnological conversion into added value products. Applied Microbiology and Biotechnology, 97(14), 6129-6147.
Liu, D., Zhang, R., Yang, X., Wu, H., Xu, D., Tang, Z., Shen, Q. 2011. Thermostable cellulase production of Aspergillus fumigatus Z5 under solid-state fermentation and its application in degradation of agricultural wastes. International Biodeterioration & Biodegradation, 65(5), 717-725.
Liu, J., Yang, J. 2007. Cellulase Production by Trichoderma koningii AS3.4262 in Solid-State Fermentation Using Lignocellulosic Waste from the Vinegar Industry. Foof Technology and Biotechnology, 45(4), 420-425.
López-González, J., Suárez-Estrella, F., Vargas-García, M., López, M., Jurado, M., Moreno, J. 2015a. Dynamics of bacterial microbiota during lignocellulosic waste composting: Studies upon its structure, functionality and biodiversity. Bioresource Technology, 175, 406-416.
López-González, J., Vargas-García, M., López, M., Suárez-Estrella, F., Jurado, M., Moreno, J. 2015b. Biodiversity and succession of mycobiota associated to agricultural lignocellulosic waste-based composting. Bioresource Technology, 187, 305-313.
Mansour, A., Arnaud, T., Lu-Chau, T., Fdz-Polanco, M., Moreira, M., Rivero, J. 2016. Review of solid state fermentation for lignocellulolytic enzyme production: challenges for environmental applications. Reviews in Environmental Science and Bio/Technology, 15(1), 31-46.
Martínez, Á., Rencoret, J., Marques, G., Gutiérrez, A., Ibarra, D., Jiménez-Barbero, J., del Río, J. 2008. Monolignol acylation and lignin structure in some nonwoody plants: A 2D NMR study. Phytochemistry, 69(16), 2831-2843.
Martins, D., Prado, H., Leite, R., Ferreira, H., Moretti, M., Silva, R., Gomes, E. 2011. Agroindustrial Wastes as Substrates for Microbial Enzymes Production and Source of Sugar for Bioethanol Production. in: Integrated Waste Management, (Ed.) S. Kumar, Vol. II, InTech.
Maulini-Duran, C., Abraham, J., Rodríguez-Pérez, S., Cerda, A., Jiménez-Peñalver, P., Gea, T., Barrena, R., Artola, A., Font, X., Sánchez, A. 2015. Gaseous emissions during the solid state fermentation of different wastes for enzyme production at pilot scale. Bioresource Technology, 179, 211-218.
Mekala, N., Singhania, R., Sukumaran, R., Pandey, A. 2008. Cellulase production under solid-state fermentation by Trichoderma reesei RUT C30: statistical optimization of process parameters. Applied Biochemistry and Biotechnology, 151(2-3), 121-131.
Mitchell, D., Berovič, M., Krieger, N. 2006. Solid-State Fermentation Bioreactor Fundamentals: Introduction and Overview. in: Solid-State Fermentation Bioreactors: Fundamentals of Design and Operation, (Eds.) D.A. Mitchell, M. Berovič, N. Krieger, Springer Berlin Heidelberg. Berlin, Heidelberg, pp. 1-12.
Chapter 1.Introduction
32
Muñiz-Márquez, D., Contreras, J., Rodríguez, R., Mussatto, S., Teixeira, J., Aguilar, C. 2016. Enhancement of fructosyltransferase and fructooligosaccharides production by A. oryzae DIA-MF in Solid-State Fermentation using aguamiel as culture medium. Bioresource Technology, 213, 276-282.
Olofsson, K., Bertilsson, M., Lidén, G. 2008. A short review on SSF – an interesting process option for ethanol production from lignocellulosic feedstock. Biotechnology for Biofuels, 1(7), 1-14.
Pagans, E., Barrena, R., Font, X., Sánchez, A. 2006. Ammonia emissions from the composting of different organic wastes. Dependency on process temperature. Chemosphere, 62(9), 1534-1542.
Pareek, N., Gillgren, T., Jönsson, L. 2013. Adsorption of proteins involved in hydrolysis of lignocellulose on lignins and hemicelluloses. Bioresource Technology, 148, 70-77.
Percival-Zhang, Y., Himmel, M., Mielenz, J. 2006. Outlook for cellulase improvement: Screening and selection strategies. Biotechnology Advances, 24(5), 452-481.
Rabinovich, M., Melnick, M., Bolobova, A. 2002. The Structure and Mechanism of Action of Cellulolytic Enzymes. Biochemistry (Moscow), 67(8), 850-871.
Ratanakhanokchai, K., Waeonukul R., Pason, P.,Tachaapaikoon, C., Kyu, K., , Sakka, K., Kosugi, A., Mori, Y 2013. Paenibacillus curdlanolyticus Strain B-6 Multienzyme Complex: A Novel System for Biomass Utilization, Biomass Now - Cultivation and Utilization, (Ed.) M.D. Matovi, InTech.
Ray, R., Behera, S. 2017. Solid State Fermentation for Production of Microbial Cellulases. in: Biotechnology of Microbial Enzymes, (Ed.) G. Brahmachari, Academic Press, pp. 43-79.
Rocha, N., Barros, M., Fischer, J., Coutinho Filho, U., Cardoso, V. 2013. Ethanol production from agroindustrial biomass using a crude enzyme complex produced by Aspergillus niger. Renewable Energy, 57, 432-435.
Rodríguez Madrera, R., Pando Bedriñana, R., Suárez Valles, B. 2015. Production and characterization of aroma compounds from apple pomace by solid-state fermentation with selected yeasts. LWT - Food Science and Technology, 64(2), 1342-1353.
Rojas, J., Bedoya, M., Ciro, Y. 2015. Current Trends in the Production of Cellulose Nanoparticles and Nanocomposites for Biomedical Applications in: Cellulose - Fundamental Aspects and Current Trends, (Ed.) M. Poletto, InTech.
Rubin, E. 2008. Genomics of cellulosic biofuels. Nature, 454(7206), 841-845. Sanghvi, G., Koyani, R., Rajput, K. 2010. Thermostable xylanase production and partial purification
by solid-state fermentation using agricultural waste wheat straw. Mycology, 1(2), 106-112. Santis-Navarro, A., Gea, T., Barrena, R., Sánchez, A. 2011. Production of lipases by solid state
fermentation using vegetable oil-refining wastes. Bioresource Technology, 102(21), 10080-10084.
Singhania, R., Sukumaran, R., Patel, A., Larroche, C., Pandey, A. 2010. Advancement and comparative profiles in the production technologies using solid-state and submerged fermentation for microbial cellulases. Enzyme and Microbial Technology, 46(7), 541-549.
Soni, R., Nazir, A., Chadha, B. 2010. Optimization of cellulase production by a versatile Aspergillus fumigatus fresenius strain (AMA) capable of efficient deinking and enzymatic hydrolysis of Solka floc and bagasse. Industrial Crops and Products, 31(2), 277-283.
Sukumaran, R., Singhania, R., Mathew, G., Pandey, A. 2009. Cellulase production using biomass feed stock and its application in lignocellulose saccharification for bio-ethanol production. Renewable Energy, 34(2), 421-424.
Sun, H., Xiangyang, G., Zhikui, H., Peng, M. 2010. Cellulase production by Trichoderma sp. on apple pomace under solid state fermentation African Journal of Biotechnology, 9(2), 163-166.
Teeri, T. 1997. Crystalline cellulose degradation: new insight into the function of cellobiohydrolases. Trends in Biotechnology, 15(5), 160-167.
Thomas, L., Larroche, C., Pandey, A. 2013. Current developments in solid-state fermentation. Biochemical Engineering Journal, 81, 146-161.
UNEP, D.C.W., L. Rodic, P. Modak, R. Soos, A. Carpintero, C.A. Velis, M. Iyer, O. Simonett, United Nations Environment Programme (UNEP) and International Solid Waste Association
Chapter 1.Introduction
33
(ISWA). 2015. United Nations Environment Programme (UNEP) and International Solid Waste Association (ISWA), (Ed.) D.B. Wilson, Global Waste Management Outlook, UNEP International Environment Technology Centre. Osaka.
Van Dyk, J., Pletschke, B. 2012. A review of lignocellulose bioconversion using enzymatic hydrolysis and synergistic cooperation between enzymes—Factors affecting enzymes, conversion and synergy. Biotechnology Advances, 30(6), 1458-1480.
Varrot, A., Frandsen, T., von Ossowski, I., Boyer, V., Cottaz, S., Driguez, H., Schülein, M., Davies, G. 2003. Structural Basis for Ligand Binding and Processivity in Cellobiohydrolase Cel6A from Humicola insolens. Structure, 11(7), 855-864.
Wang, F., Li, F., Chen, G., Liu, W. 2009. Isolation and characterization of novel cellulase genes from uncultured microorganisms in different environmental niches. Microbiol Res, 164(6), 650-7.
Wang, Z., Lu, Z., Shi, J., Xu, Z. 2016. Exploring flavour-producing core microbiota in multispecies solid-state fermentation of traditional Chinese vinegar. Scientific Reports, 6, 26818.
Wunderlin, P., Lehmann, M., Siegrist, H., Tuzson, B., Joss, A., Emmenegger, L., Mohn, J. 2013. Isotope Signatures of N2O in a Mixed Microbial Population System: Constraints on N2O Producing Pathways in Wastewater Treatment. Environmental Science & Technology, 47(3), 1339-1348.
Xiao, Z., Zhang, X., Gregg, D., Saddler, J. 2004. Effects of sugar inhibition on cellulases and beta-glucosidase during enzymatic hydrolysis of softwood substrates. Applied Biochemistry and Biotechnology, 113-116, 1115-26.
Yang, B., Dai, Z., Ding, S., Wyman, C. 2011. Enzymatic hydrolysis of cellulosic biomass. Biofuels, 2(4), 421-450.
Yang, M., Fan, D., Luo, Y., Mi, Y., Hui, J., Gao, P. 2012. Media Optimization for Cellulase Production at Low Energy Consumption with Response Surface Methodology. Energy Sources, Part A: Recovery, Utilization, and Environmental Effects, 34(20), 1883-1892.
Zverlov, V., Köck, D., Schwarz, W. 2015. The Role of Cellulose-Hydrolyzing Bacteria in the Production of Biogas from Plant Biomass. in: Microorganisms in Biorefineries, (Ed.) B. Kamm, Springer Berlin Heidelberg. Berlin, Heidelberg, pp. 335-361.
Chapter 2.Research Objectives
34
Chapter 2.Research Objectives
35
CHAPTER 2. RESEARCH OBJECTIVES
Chapter 2.Research Objectives
36
Chapter 2.Research Objectives
37
Based on the extensive background on cellulase production by SSF provided in Chapter 1, it can be
summarized that the main challenges to overcome in any SSF process are the scale-up and the
operation in a continuous configuration. For this reason, further studies must be performed on these
subjects in order to develop a more sustainable and consistent process carried out in a representative
scale.
In this sense, the main objective of this phD thesis is to develop a sustainable and reproducible
cellulase production process by means of solid-state fermentation using a low cost substrate, such as
the organic solid wastes.
In order to achieve the main objective, several specific objectives were developed and are presented
below.
• To perform a screening of several organic wastes at laboratory scale, in order to assess its
potential as cellulase source.
• To assess the scale-up of selected cellulase production processes, following cellulase
production profile and the gaseous emissions generated.
• To demonstrate the technical feasibility of a continuous solid-state fermentation focusing on
the development of operational strategies. Sequential batch operation was assessed with the
main objective to obtain a long term operation and at the same time to specialize the biomass
to degrade lignocellulosic materials. To achieve this goal operational strategies for the
production of carbohydrases: amylase and cellulase were assessed.
• To produce and to characterize an specialized a cellulose-degrading mixed inoculum. To
characterize this inoculum next generation DNA sequencing must be applied.
• To perform the cellulase production in a pilot scale solid-state fermentation using the
specialized mixed inoculum. Cellulase production profile and gaseous emissions were
followed in order to assess reproducibility.
Due to the promising results obtained during this thesis, another phD student, María Marín, is
currently working on the optimization of the enzymatic extraction step and the stabilization of the
fermented solids generated, by means of composting and anaerobic digestion processes.
Chapter 2.Research Objectives
38
Chapter 3. Materials and Methods
39
CHAPTER 3. MATERIALS AND METHODS
Chapter 3. Materials and Methods
40
Chapter 3. Materials and Methods
41
3.1. Materials: Substrates and reactors set-up
3.1.1 Substrates
The materials used during the development of the thesis were: coffee husk (CH), soy fiber (SF),
orange (OP) and lemon peels (L), almond hulls (AH), pre-cooked bread waste (BW) and olive oil
cake (OOC). All of these materials are wastes from the food industry. The substrates were obtained
from local industries in Barcelona, Spain. Coffee husk is the main waste generated by the coffee
production (Marcilla®, Mollet del Vallès, Barcelona, Spain). Specifically, coffee husk is obtained
during the drying of coffee grains. Soy fiber a by product obtained during soy milk and derivatives
production (Natursoy®, Castellterçol, Barcelona, Spain). This waste is generated as a result of the
pressing of soy grain during the process. The rest of the materials were kindly provided by local
markets. All of these materials were evaluated as substrate for cellulase production.
3.1.2 Inoculum
Compost was assessed as potential inoculum for cellulase production. This material was obtained
from the municipal solid waste treatment plant Ecoparc II (Montcada i Reixac, Barcelona, Spain).
3.1.3 Other materials
Wood chips were added into the substrate/inoculum mixture as bulking agent, in order to provide
porosity to the mixture. This material was obtained from a composting treatment plant (Manresa,
Barcelona, Spain). Specific details of all materials are provided in each chapter.
3.2 SSF reactors
3.2.1 Laboratory scale set-up
The respirometer set-up was designed and built according to Ponsá et al., (2010) and Pognani et al.,
(2011). Briefly, the respirometric system consists of 12 lines, each able to carry out 3 independent
fermentation using 0.5L reactor. These reactors were Erlenmeyer flasks with an inner system that
allowed the proper air circulation on the solid matrix. The inlet of the gas was connected to a silicone
tube that go across de reactor from top to bottom. At the bottom of the reactor and connected to the
silicone the an air diffuser is set. In addition to this system, a metallic net was set at the bottom of the
reactor with two main objectives: the first is to hold the solid material and the second is to provide
that the air freely is distributed throughout the solid. Also, each reactor is provided of one mass
airflow meters, one electro-valve that commutate the exhaust gases to CO2 and O2 detectors. The
timing used to manage the electro-valves enables to analyze the exhaust flow from each reactor
separately. A specific software was developed by GICOM research group to allow the on-line
Chapter 3. Materials and Methods
42
monitoring and the continuous storage of the experimental data (Figure 3.1). A schematic diagram of
the respirometric system is presented in Figure 3.2.
Figure 3.1. Screenshot of the software for online monitoring of different operational parameters.
Figure 3.2. Schematic diagram of respirometer used for lab-scale solid-state fermentation. Source: Pognani, (2011).
Chapter 3. Materials and Methods
43
Using this system, each reactor is able to contain a maximum of 100 g of total mixture, including
substrate, inoculum and bulking agent. Also, as the reactors are submerged in a thermostatic bath, the
temperature during the fermentations are fixed on one value. The entire screening of substrates for
cellulase production by means of solid-state fermentation was carried out using this set-up (Chapter
4).
3.2.2 Bench scale set-up
Solid-state fermentation in bench scale were performed in either 4.5 L or 10 L air-tight packed-bed
reactors, thermally isolated to work under near adiabatic conditions. Experimental set-up is presented
in Figure 3.3. Air was continuously supplied to the reactors by means of a mass airflow controller
(Bronkhorst, Spain). Airflow was automatically adjusted by a feedback controller in order to maintain
oxygen content in the reactor in levels above 5% and avoid anaerobic conditions. In this context, air
flow was supplied through a pipeline from the bottom of the reactor where, by means of a plastic
diffusor, the air circulated through the solid bed of the reactor. The exit of the exhausted gas is located
on the top of the reactor, and led to a oxygen analyzer. The oxygen content was measured using a
oxygen sensor (Sensortran, Spain) in a range of (0-20.9%).
Between the reactor and the oxygen sensor there is a vapor-condensing device to avoid humidity to
reach the sensor and consequently damage. All outlet pipelines were connected to a biofilter.
Figure 3.3. Experimental set-up of the bench scale solid-state fermentation system.
During all fermentations airflow, temperature and oxygen content were continuously monitored.
Monitorization was performed by a self-made acquisition and control system based on Arduino® and
self-made software. The control system is fully described in section 3.4 of this chapter.
Chapter 3. Materials and Methods
44
3.2.3 Pilot scale set-up
This system is located outside of the Chemical, Biological and Environmental Engineering
Department at UAB. It consists of 2 reactors of 55 L total volume of which 50 L is the working
volume (shown in Figure 3.3). These reactors were used in all experiments at pilot scale.
The system was supplied with filtered compressed air provided by a pressure compressor Decibar 30
Worthington®. Two connectors were place at the cover of both reactors. One of the connectors was
designed to hold and insert the temperature probe and the other one to insert the outlet pipeline of
exhausted gases. Another pipeline is set at the bottom of the reactor, which allowed the injection of
air. Exhausted gases exiting from the top of the reactor firstly passed through a water trap and then
went to the O2 sensor (Xgard 501/265/S, Crowcon, England).
During all fermentations airflow, temperature and oxygen content were continuously monitored.
Monitorization was performed by a self-made acquisition and control system based on Arduino® and
self-made software. The control system is fully described in section 3.4 of this chapter.
Figure 3.5. Picture of pilot reactors of 50-L of working volume.
3.3 Specific methods for the monitoring of cellulase production process.
3.3.1 Specific Oxygen Uptake Rate and Cumulative Oxygen Consumption.
Specific oxygen uptake rate (sOUR) was calculated on-line for continuous monitoring in order to
provide more accurate information on biological activity, according to:
Chapter 3. Materials and Methods
45
3
3
10··10·6032)209.0(
2 DWTRPyFsOUR O ⋅
⋅⋅⋅−⋅= Equation (1)
where: sOUR is the specific Oxygen Uptake Rate (mg O2 g-1 DM h-1); F, airflow (mL min-1); yO2, is
the oxygen molar fraction in the exhaust gases (mol O2 mol-1); P, pressure of the system assumed
constant at 101325 Pa; 32, oxygen molecular weight (g O2 mol O2-1); 60, conversion factor from
minute to hour; 103, conversion factor from mL to L; R, ideal gas constant (8310 Pa L K-1 mol-1); T,
temperature at which F is measured (K); DW, initial dry weight of solids in the reactor (g); 103,
conversion factor from g to mg. Moreover, the area below the sOUR curve was also determined,
which represents the cumulative oxygen consumption (COC) during the process (Ponsá et al., 2010).
These parameters will provide information on the overall biological activity and one of the main
energy requirements for cellulases production.
3.3.2 Enzyme extraction
Cellulases were extracted by adding 150 mL of citrate buffer (0.05 M, pH 4.8) to 10 g of fermented
solids in a 250 mL Erlenmeyer flask and mixing thoroughly on a magnetic stirrer for 30 min at room
temperature. The mixture was separated by centrifugation at 10000 rpm for 10 min, followed by
filtration with a 0.45 µm filter. The remaining supernatant was used for cellulase activity
determination (Dhillon et al., 2012).
3.3.3 Cellulase activity
The components of the cellulase system were measured in terms of carboxymethyl cellulase
(CMCase), filter paper activity (FPase) and β -glucosidase (BGase) according to the methods
recommended by IUPAC (Ghose, 1987).
FPase is the most accepted method to reflect total cellulase action. The main advantage of this
procedure is that requires simple reagents and equipments; however, it has been reported as a very
laborious and time-consuming method, requiring many manual manipulations which lead to non-
reproducible results (Coward-Kelly et al., 2003). Also, it is important to consider possible mass
transfer problems when filter paper is used and possible non adequate contact surface of the paper
strips in the reaction medium. In spite of this considerations, until this day remains as the most
utilized method to describe cellulase activity. For these reasons microcrystalline cellulose (SC) was
Chapter 3. Materials and Methods
46
evaluated as a possible substrate to replace the filter paper; however, for the difficulties to make
proper comparison with reported studies, filter paper was used during the entire thesis.. These results
are shown in Appendix III.
FPase and CMCase assay are based on the reducing sugars measurements. The final products were
measured using dinitrosalicylic acid (DNS) reagent (Miller, 1959). In the case of BGase, the final
product obtained is glucose, this was measured using a YSI 2700 Select Biochemistry Analyser (John
Morris Scientific, Perth). One unit of the CMCase (U), FPase (FPU) and BGase (U) were expressed as
being equivalent to the amount of enzyme that releases 1µmol of glucose from CMC, Whatman filter
paper and cellobiose respectively in 0.05 M citrate buffer, pH 4.8 at the determined time of assay.
In all cases, enzymatic activity production has been expressed with respect to the dry matter content,
i.e, FPU g-1 DM for FPase and U g-1 DM for CMCase and BGase.
Reagents
Citrate Buffer: All cellulase assays were carried out in 0.05 M citrate buffer pH 4.8. This buffer was
made by adding 210 g of citric acid monohydrate (C6 H8 07 * H20) to 750 mL distilled water and mix.
Then, NaOH was added until pH reached 4.3 and diluted to 1000 mL and check pH. If necessary,
NaOH until pH equals 4.5. This is 1 M citrate buffer at pH 4.5. When diluted to 0.05 M, pH should be
4.8-5.
DNS Reagent: Prepare a solution adding 10.6 g of 3,5-dinitrosalicylic acid (DNS) and 19.8 g of
NaOH to 1416 mL of distilled water and mix. Once the solution is fully dissolved add 306 g of
sodium potassium tartrate, 7.6 mL of phenol (if phenol is solid, melt at 50°C) and 8.3 g sodium
metabisulfite, then mix.
Procedure
All of the following procedures are described as reported by Ghose, (1987).
a) Filter Paper Activity Assay (FPase).
Substrate: Filter paper strips (6 x 1 x 1 cms).
Procedure:
- Add 1.0 ml 0.05 M Na-citrate buffer, pH 4.8 and 0.5 mL of properly diluted enzyme, to a test tube of
volume at least 25 ml. Add one filter paper strip and mix.
Chapter 3. Materials and Methods
47
- Along with the samples, prepare controls for substrate and enzyme. Substrate control is prepared by
replacing the enzyme volume for distilled water. Enzyme control is prepared by replacing substrate
volume for distilled water. All controls and samples must be prepared at least in triplicates.
- Incubate all controls and samples at 50°C for 60 min.
- Add 3.0 mL DNS and mix. Boil for exactly 5.0 min in a vigorously boiling water bath containing
sufficient water. All samples, controls and the spectro zero should be boiled together.
- After boiling, transfer to a cold water bath. Add 20 mL deionized or distilled water. Mix by
completely inverting the tube several times so that the solution separates from the bottom of the tube
at each inversion.
- Measure against the spectro zero at 540 nm.
Calibration curve:
Substrate volume is replaced by deionized or distilled water. Enzyme volume is replaced with glucose
standards at several different concentrations. Then, follow the same procedure that for activity
measurement. An example of the obtained calibration curves is presented in Appendix I.
Calculations
The calculations were carried out using the following equation:
𝐹𝑃𝑎𝑠𝑒 (𝐹𝑃𝑈 𝑔!!𝐷𝑀) = !!"#$%& ∗!∗!∗!""!.!"∗!"∗!"
(Equation 2)
where, Csample is the concentration of reducing sugars in g L-1 , D is the dilution factor of the enzymatic
extract, E is the extraction factor in (g of fermented solid/ mL of buffer) and DM is the dry matter
content of the fermented solid.
b) Carboxymethylcellulase activity Assay (CMCase).
Substrate: Carboxymethylcellulose 2% prepared using 0.05 M Na-citrate buffer at pH 4.8.
Procedure:
- Add 0.5 ml of carboxymethylcellulose solution and 0.5 mL of properly diluted enzyme extract, to a
test tube of volume at least 25 ml and mix.
Chapter 3. Materials and Methods
48
- Along with the samples, prepare controls for substrate and enzyme. Substrate control is prepared by
replacing the enzyme volume for distilled water. Enzyme control is prepared by replacing substrate
volume for distilled water. All controls and samples must be prepared at least in triplicates.
- Incubate all controls and samples at 50°C for 60 min.
- Add 3.0 mL DNS and mix. Boil for exactly 5.0 min in a vigorously boiling water bath containing
sufficient water. All samples, controls and the spectro zero should be boiled together. After boiling,
transfer to a cold water bath. Add 20 ml deionized or distilled water. Mix by completely inverting the
tube several times so that the solution separates from the bottom of the tube at each inversion.
- Measure against the spectro zero at 540 nm.
Calibration curve:
Substrate volume is replaced by deionized or distilled water. Enzyme volume is replaced with glucose
standards at several different concentrations. Then, follow the same procedure that for activity
measurement. An example of the obtained calibration curves is presented in Appendix II.
Calculations
The calculations were carried out using the following equation:
𝐶𝑀𝐶𝑎𝑠𝑒 (𝑈 𝑔!!𝐷𝑀) = !!"#$%& ∗!∗!∗!""!.!"∗!"∗!"
(Equation 3)
where, Csample is the concentracion of reducing sugars in g L-1 , D is the dilution factor of the
enzymatic extract, E is the extraction factor in (g of fermented solid/ mL of buffer) and DM is the dry
matter content of the fermented solid.
c) β-glucosidase activity Assay (BGase)
Substrate: Cellobiose 15mM prepared using 0.05 M Na-citrate buffer at pH 4.8.
Procedure:
- Add 1.0 ml of cellobiose solution and 1.0 mL of properly diluted enzyme extract, to a test tube of
volume at least 25 ml and mix.
- Along with the samples, prepare controls for substrate and enzyme. Substrate control is prepared by
replacing the enzyme volume for distilled water. Enzyme control is prepared by replacing substrate
volume for distilled water. All controls and samples must be prepared at least in triplicates.
Chapter 3. Materials and Methods
49
- Incubate all controls and samples at 50°C for 30 min.
- Transfer the tube to a cold water bath and determine glucose produced using YSI 2700 Select
Biochemistry Analyser.
Calculations
The calculations were carried out using the following equation:
𝐵𝐺𝑎𝑠𝑒 (𝑈 𝑔!!𝐷𝑀) = !!"#$%& ∗!∗!∗!""!.!"∗!"∗!"
Equation (4)
where, Csample is the concentracion of reducing sugars in g L-1 , D is the dilution factor of the
enzymatic extract, E is the extraction factor in (g of fermented solid/ mL of buffer) and DM is the dry
matter content of the fermented solid.
3.3.4 Microbial characterization
Identification of the microbial population was performed in several solid samples, obtained from
fermentation processes, using next generation sequencing. The aim of this analysis was to determine
the potential variation on biodiversity and to characterize biomass obtained during enzyme
production.
Total DNA was extracted and purified using PowerSoil™ DNA Isolation Kit (MoBio Laboratories,
USA) according to provider’s specifications. DNA samples were checked for concentration and
quality using the NanoDrop ND-1000 Spectrophotometer (NanoDrop Technologies, Wilmington,
Delaware; USA) (López-González et al., 2014).
Bacterial 16S rRNA hypervariable regions V3-V4 and Fungal ITS1-ITS3 were targeted. Later
sequencing of the extracted DNA and bioinformatics were performed on MiSeq platform by Life
Sequencing S.A (Valencia, Spain).
3.3.5. Gaseous emissions determination
All gaseous emissions measurements were developed according to the reported by Colón et al., (2012)
and volatile organic compounds determination was carried out as described by Maulini-Durán et al.,
(2014).
a) Ammonia determination
Chapter 3. Materials and Methods
50
Ammonia measurements were performed using an ammonia sensor iTX T82 (Oakdale, PA, USA),
which have a range of measurement of 0-200 ppmv. The sensor has to be set inside an hermetically
sealed recipient. This recipient has two valves to serve as gas inlet and outlet, this system allows the
proper circulation of the airflow. This system has to be located at the gas outlet of the reactor, before
the water trap. Ammonia measurement is obtained once the sensor stops fluctuating.
b) Methane, nitrous oxide and volatile organic compounds determination
Conditions
As for methane (CH4) and nitrous oxide (N2O) determination, these measurements were carried out by
gas chromatography. Results are expressed in parts per million in volume (ppmv). Details of the
equipment and operational conditions for the analysis is provided in Table 3.1 and Table 3.2 for
methane and nitrous oxide, respectively.
Table 3.1. Specifications of the equipment and conditions of analysis for methane determination.
Equipment
Cromatograph Agilent Tech. 6890N GC System
Column HP-PLOT 30m·0.53mm·40µm (Agilent Tech.)
Detector Flame ionization detector (FID)
Conditions
Injector temperature (ºC) 240
Detector temperature (ºC) 250
Oven Temperature (ºC) Isothermal process at 60ºC
Carrier gas Nitrogen at 4 psi and split 1:2
Time of analysis (min) 4
Injection volume (µL) 500
VOC analysis was performed by gas chromatography at specific conditions, which are presented in
Table 3.3. Quantification of these gases was carried out including all organic compounds present on
the gaseous stream, including methane content, but without any further identification. Then, the total
VOC value is obtained by subtracting the methane content to the total of organic compounds detected
by the chromatograph. VOC concentration is obtained in units of concentration expressed as
milligrams of total carbonaceous compounds per cubic meters of gas (mg Ctotal/m3).
Chapter 3. Materials and Methods
51
Calibration curves
The calibration is carried out by using a standard gas with known concentration, in order to determine
the response of the method and therefore, to be able to construct a calibration curve. The used
standard gases were n-hexane (Spigno et al., 2003), methane and nitrous oxide (purity of 99.99%) for
VOC. methane and nitrous oxide, respectively.
Calculations
The chromatograph provide de measurement on the gases concentration integrating the area obtained
in each injection by means of ChromeleonTM Software.
Table 3.2. Specifications of the equipment and conditions of analysis for nitrous oxide determination.
Equipment
Cromatograph Agilent Tech. 6890N GC System
Column HP-PLOT 30m·0.53mm·40µm (Agilent Tech.)
Detector Electron capture detector (ECD)
Conditions
Injector temperature (ºC) 120
Detector temperature (ºC) 345
Oven Temperature (ºC) Isothermal process at 60ºC
Carrier gas Nitrogen at 4 psi and split 1:2
Time of analysis (min) 4
Injection volume (µL) 500
c) VOC identification and quantification
A sample from each process was taken in a 250 mL glass gas collector. VOC characterization was
performed using air samples analyzed by SPME (Solid Phase Micro Extraction)/GC–MS, as
previously reported by other authors (Orzi et al., 2010). A manual SPME device with divinylbenzene
(DVB)/Carboxen/ polydimethylsiloxane (PDMS) 50–30 µ m fiber from Supelco (Bellefonte, PA,
USA) was used. The compounds were adsorbed from the air samples by exposing the fiber
(preconditioned for 1 h at 270 C, as suggested by the supplier) to the sample in the glass gas collector
for 30 min at room temperature. A solution of deuterated p-xylene in methanol was used as internal
standard (IS). VOC characterization was performed using a Gas Chromatograph (Agilent 5975C)
Chapter 3. Materials and Methods
52
coupled with a 7890 Series GC/MSD. Volatile compounds were separated using a capillary column
for VOC (Agilent Technologies DB-624) measuring 60 m x 0.25 mm with a film thickness of 1.40
µm. Carrier gas was helium at a flow rate of 0.8 mL min-1. VOC were desorbed by exposing the fiber
in the GC injection port for 3 min at 250ºC. A 0.75 mm internal diameter glass liner was used, and the
injection port was in splitless mode
Table 3.3. Specifications of the equipment and conditions of analysis for volatile organic compounds
determination.
Equipment
Cromatograph Agilent Tech. 6890N GC System
Column Tracsil TRB-1, 2m·0.53mm·3µm (Teknokroma)
Detector Flame ionization detector (FID)
Conditions
Injector temperature (ºC) 250
Detector temperature (ºC) 250
Oven Temperature (ºC) Isothermal process at 200ºC
Carrier gas Helium at 1.5 psi and splitless
Time of analysis (min) 1
Injection volume (µL) 250s
The temperature program was isothermal for 2 min at 50ºC, raised to 170ºC at a rate of 3ºC min-1
and, finally, to 230ºC at a rate of 8ºC min-1 . The transfer line to the mass spectrometer was
maintained at 235ºC. The mass spectra were obtained by electron ionization at 70 eV, a multiplier
voltage of 1379 V and collecting data over the mass range of 33–300. Deuterated p-xylene was used
to determine the fiber and GC– MS response factors for 15 typical compounds emitted in composting
processes according to the literature (Scaglia et al., 2011)These 15 compounds were diluted in
methanol at the same concentration as deuterated p-xylene. This solution (10 µL) was injected into the
glass gas collector with 10 µL of deuterated p-xylene in methanol solution. The fiber was exposed for
30 min to the resulting solution and injected into the GC–MS using the same method as described
above. The area obtained for each compound was compared to deuterated p-xylene to determine each
response factor. The aim of determining these response factors is to increase the reliability of the
quantitative analysis. Compounds were identified by comparing their mass spectra with the mass
spectra contained in the NIST (USA) 98 library. A semi-quantitative analysis for all the identified
compounds was performed by direct comparison with the internal standard. Quantitative analysis was
Chapter 3. Materials and Methods
53
performed for m-xylene, n-decane, alphapinene, beta-pinene, limonene, toluene, dimethyl disulfide,
hexanal, styrene, cyclohexanone, nonanal, decanal, eucalyptol, pyridine and 2-pentanone. These
compounds have been the most common VOC found in previous experiments (Maulini-Duran et al.,
2014), representing different VOC families.
3.4 Aeration control system
Two different control systems were used during the development of this thesis: oxygen feedback
control and sOUR control.
3.4.1 Oxygen feedback control
The controller was based on the airflow manipulation by means of the oxygen content measured in the
exhaust gas. It was necessary to establish an O2 set point to maintain the system in favorable
conditions. The oxygen set point was fixed between 11.5 and 12.5% (v/v) and considered as the
optimum oxygen content . Emulating the controllers used at industrial level, the controller applied a
high airflow for oxygen levels below 11.5% and a low airflow for measures over 12.5%, whereas the
controller did not take action when the measure was within this range. The highest and lowest
airflows depended on the dynamic of every assessed raw material and the type of reactor (0.5, 4.5, 10
or 50 L). Because of the system's slow dynamics the closed loop applied was set up to work in cycles
of 15 min.
3.4.2 Oxygen Uptake Rate (sOUR) control
In this case, the measured variable is the sOUR (determined using the online measured oxygen
content) and the manipulated variable is the airflow. The sOUR is a parameter that provides direct
information on the biological activity of the reactors, it is for this reason that the main goal of this
control is to maximize this parameter. More details on the development of this control strategy is
published by Puyuelo et al., (2010). A full schematic diagram of the control system is presented in
Figure 3.4.
Briefly, the maximization of the sOUR was achieved through a control system working in cycles of
one hour each, according to the residence time distribution study. After completing a cycle, the
oxygen level is revised to avoid percentages below 5% (v/v). If the level of oxygen is less than the
limit, the airflow will be increased 50%. If an adequate oxygen level is detected, the system will start
to control the specific oxygen uptake rate (sOUR) measurements. Hence, values of flow and sOUR,
which are related, are re-evaluated by the controller between two consecutive cycles. For both
Chapter 3. Materials and Methods
54
parameters, three situations are possible, for instance, the system determines if the current value is
lower than, higher than or equal to the previous value. Different absolute thresholds were established
to defined the superior and inferior limits in which the variation of sOUR and airflow can be
considered negligible. The limit to detect sOUR variation was defined as 0.5 % of the maximum
sOUR by previous studies. The range considered for the airflow measurements was 0.05 L min-1.
Figure 3.4. Scheme of control laws for the OUR controller. F: airflow; OUR: oxygen uptake rate.
Source: Puyuelo et al., (2010).
3.5 Standard methods
Routine methods were determined according to standard procedures included in the "Test Methods for
the Examination of Composting and Compost" (US Department of Agriculture and US Composting
Council, 2001). If it is not included in this reference, the original reference will be provided. All
results are calculated using triplicates in all cases.
Chapter 3. Materials and Methods
55
3.5.1 Moisture and Dry matter content.
To determine moisture (M) and dry matter content (DM, equivalent to total solids), a determined
amount of solid sample is set on a previously weighted dry capsule and dried in an oven at 105ºC
during 24h, then weighted and calculated according to the following equations:
( ) ( )( )
( ) MDM
WoWiWfWiM
%100%
100*%
−=
−−
=
(Equation 5)
where Wo is the weight of the dry capsule, Wi is the initial wet weight of the sample and Wf is the
final dry weight of the sample.
3.5.2 Organic matter content.
To determine the organic matter content (OM, equivalent to volatile solids), the dried solid sample is
submitted to ignition at 550ºC during 4h and it is calculated as follows:
( ) ( )( )
100*%WoWiWaWiM
−−
= (Equation 6)
where, Wo is the weight of the dry capsule, Wi is the initial dry weight of the sample and Wa is the
final weight of the ashes, obtained after ignition.
3.5.3 pH
pH was determined by mixing a ratio of 1:5 w/v of sample into distilled water. The sample was
shaken at room temperature during 30 min to allow the salts to solubilize into the liquid phase. Then
centrifuge at 3500 rpm during 10 min and measured. pH was measured using an electronic pH-meter
(Crinsom®, micro CM2100).
3.5.4 Fiber content.
The fibers content were determined using the method reported by Van Soest et al. (1991). According
to these authors, cellulose content is the result of the difference between neutro detergent fibers
(NDF) and acid detergent fibers (ADF). Also, hemicellulose content is corresponds to the difference
between ADF and lignin content (ADL).
Chapter 3. Materials and Methods
56
a) Neutro detergent fibers (NDF)
Its determination is based on the solubility of the fiber components to dodecyl sodium sulphate at
neutral pH. The soluble components of the cell wall such as starch and simple sugars are solubilized
by the detergent, while recalcitrant components such as cellulose, hemicellulose and lignin remain in
the solid matrix.
Reagent: NDF reagent was made using 1.86 g EDTA, 0.68 g Na2B4O7·10H2O, 3g sodium lauryl
sulphate, 1 g of 2-etoxyethanol, 0.456 g of Na2HPO4 and 95 mL of H2O. Then adjusted pH to 7.0
using H3PO4.
Procedure:
- Weight 1 g of wet sample and add 100 mL of NDF reagent, 2mL of decahydronaphthalene and 0.5
mL (1.316 g) of Na2SO3
-Autoclave for 15 minutes at 121ºC.
- Filtrate using 0.45 µm glass fiber filter (previously dried and weighted).
- Wash with 500 mL of hot water, then distilled water and finally with ketone, in order to remove
pigments and water residues.
- Dry during 4 h at 100ºC. When the sample is temperate, weight.
Calculations
The calculations were carried out using the following equation:
( ) ( )( )
ADFNDFCellulose
WiWfWiNDF
−=
−=
(%)
100*%
(Equation 7)
where, Wi is the initial weight of the wet sample and Wf is the final dry sample.
b) Acid detergent fibers (ADF)
Its determination is based on the cellular components solubility in a cethylmethyl ammonium bromide
solution. The residue obtained after the treatment contains mainly cellulose and lignin.
Chapter 3. Materials and Methods
57
Reagent: ADF reagent was made my mixing 2 g of cethylmethyl ammonium bromide and 100 mL
H2SO4 1 N.
Procedure:
- Weight 1 g of wet sample in an Erlenmeyer flask and add 100 mL of ADF reagent and 2 mL of
decahydronaphthalene.
- Autoclave for 15 minutes at 121ºC.
- Filtrate using 0.45 µm glass fiber filter (previously dried and weighted).
- Wash with 500 mL of hot water, then distilled water and finally with ketone, in order to remove
pigments and water residues.
- Dry during 4 h at 100ºC. When the sample is temperate, weight.
Calculations
The calculations were carried out using the following equation:
( ) ( )( )
ADLADFoseHemicellul
WiWfWiADF
−=
−=
(%)
100*%
(Equation 8)
where, Wi is the initial weight of the wet sample and Wf is the final dry sample.
c) Lignin (ADL)
- The dry solid obtained as result for ADF determination is used for the ADL determination.
- Cover the solid with sulfuric acid 72% and wait for 1 h. Remove the acid and repite 3 times.
- Filtrate using 0.45 µm glass fiber filter (previously dried and weighted) and wash with 500 mL of
hot water.
- Dry until constants weight at 105ºC (W1).
- Submit the dried sample to ignition at 550ºC and weight (W2).
Calculations
The calculations were carried out using the following equation:
Chapter 3. Materials and Methods
58
( ) ( )( )
ADLLignin
WiWWADL
=
−=
(%)
100*% 21
(Equation 9)
where, Wi is the initial weight of the wet sample and Wf is the final dry sample.
References
Colon, J., Cadena, E., Pognani, M., Barrena, R., Sanchez, A., Font, X., Artola, A. 2012. Determination of the energy and environmental burdens associated with the biological treatment of source-separated Municipal Solid Wastes. Energy & Environmental Science, 5(2), 5731-5741.
Coward-Kelly, G., Aiello-Mazzari, C., Kim, S., Granda, C., Holtzapple, M. 2003. Suggested improvements to the standard filter paper assay used to measure cellulase activity. Biotechnology and Bioengineering, 82(6), 745-749.
Dhillon, G., Kaur, S., Brar, S., Verma, M. 2012. Potential of apple pomace as a solid substrate for fungal cellulase and hemicellulase bioproduction through solid-state fermentation. Industrial Crops and Products, 38(1), 6-13.
Ghose, T. 1987. Measurement of cellulase activities. Pure and Appied Chemistry, 59(2), 257-268. López-González, J., Vargas-García, M., López, M., Suárez-Estrella, F., Jurado, M., Moreno, J. 2014.
Enzymatic characterization of microbial isolates from lignocellulose waste composting: Chronological evolution. Journal of Environmental Management, 145, 137-146.
Maulini-Duran, C., Artola, A., Font, X., Sánchez, A. 2014. Gaseous emissions in municipal wastes composting: Effect of the bulking agent. Bioresource Technology, 172, 260-268.
Miller, G. 1959. Use of Dinitrosalicylic Acid Reagent for Determination of Reducing Sugar. Analytical Chemistry, 31(3), 426-428.
Pognani, M. 2011. Organic matter evolution in biological solid-waste treatment plants. Raw waste and final product characterization. in: Chemical Engineering Department, Vol. PhD, Universitat Autònoma de Barcelona. Barcelona.
Pognani, M., Barrena, R., Font, X., Adani, F., Scaglia, B., Sánchez, A. 2011. Evolution of organic matter in a full-scale composting plant for the treatment of sewage sludge and biowaste by respiration techniques and pyrolysis-GC/MS. Bioresource Technology, 102(6), 4536-4543.
Ponsá, S. 2010. Different indices to express biodegradability in organic solid wastes: Application to full scale waste treatment plants. in: Chemical Engineering Department, Vol. PhD Universitat Autònoma de Barcelona. Barcelona, Spain.
Ponsá, S., Gea, T., Sánchez, A. 2010. Different Indices to Express Biodegradability in Organic Solid Wastes Journal of Environmental Quality, 39(2), 706-712.
Puyuelo, B., Gea, T., Sánchez, A. 2010. A new control strategy for the composting process based on the oxygen uptake rate. Chemical Engineering Journal, 165(1), 161-169.
Scaglia, B., Orzi, V., Artola, A., Font, X., Davoli, E., Sanchez, A., Adani, F. 2011. Odours and volatile organic compounds emitted from municipal solid waste at different stage of decomposition and relationship with biological stability. Bioresource Technology, 102(7), 4638-4645.
Spigno, G., Pagella, C., Daria Fumi, M., Molteni, R., Marco De Faveri, D. 2003. VOCs removal from waste gases: gas-phase bioreactor for the abatement of hexane by Aspergillus niger. Chemical Engineering Science, 58(3–6), 739-746.
Chapter 3. Materials and Methods
59
Van Soest, P., Robertson, J., Lewis, B. 1991. Methods for dietary fiber, neutral detergent fiber, and non starch polysaccharides in relation to animal nutrition. Journal of Dairy Science, 74(10), 3583-3597.
Chapter 4. Screening of wastes
60
Chapter 4. Screening of wastes
61
CHAPTER 4. CELLULASE PRODUCTION BY
SOLID-STATE FERMENTATION USING
ORGANIC WASTES AY LAB SCALE AND
SCALE-UP.
Part of this chapter is published in: Maulini-Duran, C., Abraham, J., Rodríguez-Pérez, S., Cerda, A.,
Jiménez-Peñalver, P., Gea, T., Barrena R., Artola, A, Font, X., Sánchez, A. 2015. Gaseous emissions
during the solid state fermentation of different wastes for enzyme production at pilot scale.
Bioresource Technology: 179, 211-218.
Chapter 4. Screening of wastes
62
Chapter 4. Screening of wastes
63
Introduction
In this chapter four different residues provided by several local industries were screened in order to
determine its suitability for cellulase production at lab scale. The raw materials that provided highest
cellulase production, coffee husk and orange peels, were assessed as substrates at a larger scale using
10-L reactors (3-4 Kg). The main objective of working at this scale is to submit the solid mixture to a
self-heating process and to assessed the scalability effect when compared with the previous screening.
A final scale-up of these processes were carried out in a pilot scale, working in a 50-L reactor (15-20
Kg). In these experiments, in addition to the cellulase production profile, the released gaseous
emissions were measured in order to partially assess the potential environmental impact of the SSF.
The experiments showed in this chapter were carried out as a joint venture with Pedro Jimenez and in
collaboration with Caterina Maulini.
4.1 Methodology
4.1.1 Screening of raw materials for cellulase production
A screening of several organic wastes was performed in order to determine their potential for cellulase
production. Orange (OP) and lemon (L) peels, olive oil cake (OOC) and coffee husk (CH) were
analyzed. Also, wood chips (WC) and almond hull (AH) were assessed as alternative bulking agents.
A full characterization of all the materials is presented in Table 4.1. Compost (C) was added in all
cases as inoculum in order to provide microbial diversity to the solid mixture (López-Gonzalez et al.,
2015b).
The fermentations were carried out using a mixture comprising each waste as substrate and compost
as inoculum in a 90:10 (w/w) ratio. In addition bulking agent was added to provide porosity and
improve the oxygen transfer in the solid bed. Each fermentation was performed in triplicates during 4
days using the experimental set-up described in section 3.1.4. Operational conditions of the
fermentations were constant, with a temperature of 37ºC and airflow of 20 mL min-1. Two samplings
were performed, one at the moment of maximum sOUR and another at the end of the fermentation.
Cellulase activities of FPase, CMCase and BGase were measured in all samples taken from the
reactors. In addition to the cellulase activities and sOUR monitoring, several physical-chemical
parameters were measured: pH, dry matter, organic matter, reducing sugars and glucose.
Chapter 4. Screening of wastes
64
Table 4.5. Characterization of all evaluated wastes. (*) Mean value ± standard deviation (n=3), (**)s.d< 6% n.m: not measured.
Parameter OP L OOC AH CH C
pH (1:5) 4.64±0.02 5.4±0.01 4.50±0.01 5.96±0.01 6.5±0.01 7.96±0.02
CE (mS cm-1) 0.67±0.01 0.85±0.01 2.38±0.10 0.49±0.01 n.m 6.25±0.20
Wet basis(%)
Moisture 82.3±0.2 80.75±0.04 75.7±0.3 12.9±0.3 59.9±0.4 34.29±0.01
Dry matter 17.7±0.2 19.25±0.04 24.3±0.3 87.1±0.3 40.1±0.4 64.71±0.01
Dry basis (%)
Organic matter 98±2 95.85±0.01 95.2±0.3 99.01±0.37 90.22±0.08 n.m
Reducing sugars 37±1 36.3±0.2 2.51±0.01 0.5±0.01 0.65±0.01 n.m
Glucose 6±0.01 6±0.01 0.29±0.01 0.08±0.01 0.02±0.01 n.m
Total C** 44.5 55 48 80.1 n.m
Total H** 6.2 7.6 6 n.m n.m
Total N** 1.01 2 0.5 3.5 n.m
C/N ratio 44 27.5 96 22.9 n.m
4.1.2. SSF scale-up assessment
From the screening described in the previous section, two raw materials were selected for further
assessment of the process scale up.
In all the fermentations performed in near-adiabatic reactors, the selected wastes were evaluated as
substrates using compost as inoculum in a 90:10 (w/w) ratio. Also the selected bulking agent was
added in a 1:1 (v:v) ratio to each fermentation. In a first stage, the fermentations were carried out
using the system described in section 3.1.5 using 10-L reactors. The system was operated using the
sOUR control as described in section 3.4.2 starting with an initial airflow of 100 mL min-1. During the
fermentations, solid samples were taken starting at the moment of maximum sOUR to a final of 6
sampling, in order to determine the cellulase production profile and the specific substrate profile,
reflected as reducing sugars and glucose.
Chapter 4. Screening of wastes
65
In addition to the 10-L SSF, the selected wastes were assessed as raw materials in a pilot scale, using
the system described in section 3.1.6 and performed in duplicates. The system was operated also using
the sOUR control as described in section 3.4.2 starting with an initial airflow of 200 mL min-1. In this
case, also the substrate : inoculum ratio was 90:10 (w/w) and the bulking agent ratio was 1:1 (v/v).
Cellulase activity, reducing sugars and glucose profiles were measured during the sampling. Also,
total gaseous emissions of CH4, N2O, VOCs and NH3 were measured. The samples were collected
from reactors’ outlet in 1 L Tedlar® bags and determined by gas chromatography (Agilent
Technologies 6890N Network GC system, Madrid, Spain). The composition of the VOCs emissions
was also determined and quantified. Both of these procedures were carried out according to the
method described by Colon et al., (2012) (Chapter 3, section3.3.5).
4.2 Results
4.2.1 Screening of raw materials for cellulase production
Maximum biological activity detected in the fermentations is presented in Table 4.2 expressed as
sOUR. It is possible to observe that a moderate (to high) biological activity was obtained using all the
different substrates. The main differences can be found when is considered the characterization of the
raw materials or even the initial mixture. It this sense, the residues coming from a direct agricultural
source such as the orange and lemon peel presented a high content of readily metabolizable
compounds such as reducing sugars and glucose as seen in Table 4.1. In contrast, coffee husk
presented the lowest sOUR and also showed the lowest content of soluble materials. Then, it is
possible that the high sOUR values obtained for orange and lemon peels and also olive oil cake are
related to the initial content of soluble components.
In Table 4.2 is also presented the time when the maximum sOUR was obtained. It is clearly observed
that when using orange peels as substrate and wood chips as bulking agent the fermentation time is
longer than in the rest of experiments. This can be due to the initial acidic pH of the mixture (pH=
4.9), which could delay the starting up of the process as reported by Sundberg et al., (2013).
In contrast, when almond hulls are used as bulking agent, the time of fermentation decreases, reaching
the maximum sOUR in only 1 day. This fact can be attributed also to the pH. In this case, the initial
mixture had a initial pH of 5.5, which is significantly higher that when using wood chips as bulking
agent. Almond hulls used as a bulking agent presented positive results, in terms of biological (Table
4.2) and enzymatic activity, which can be due to the high antioxidant content and the cell-protective
characteristics of this residue (Moosavi Dolatabadi et al., 2015).
Chapter 4. Screening of wastes
66
Table 4.6. sOUR values and time of fermentation obtained at the moment of maximum cellulase activity, working at lab scale at 37ºC.
Waste sOUR (mgO2 g-1DM h-1) Time (d)
Orange peel + Wood chips 3.5 ± 0.3 2
Orange peel + Almond hull 2.3 ± 0.1 1
Lemon peel + Almond hull 2.5 ± 0.1 1
Coffee husk + Wood chips 1.6± 0.1 1
Olive oil cake +Wood chips 3.1 ± 0.1 1
During the process it was observed a rapid reduction until negligible values of the reducing sugars and
glucose (data not presented). The only exception relies on the experiment carried out with orange
peels and almond hulls, where an accumulation of nearly 30% of glucose occurred. It has to be
considered that reducing sugars and glucose content in the fermentation represents a net value of the
process, which includes hydrolysis of polysaccharides and consumption of the released sugars. It was
expected that most of the glucose would be immediately consumed by microorganisms, however in
OP and AH experiments an accumulation this monosaccharide took place, which can be of interest
considering further studies regarding bioethanol production.
Figure 4.2 presents the three cellulase activity values obtained in all experiments. It is clearly
observed that highest FPase activity was achieved by the fermentation using as the substrate coffee
husk followed by orange peels, with 6.4±0.7 and 1.6±0.5 FPU g-1DM, respectively. The rest of the
assessed substrates obtained values below the unit of activity. Orange peels is commonly used as s
substrate for pectinase or pectine derivates production for its availability and composition (Grohmann
& Baldwin, 1992; Li et al., 2016), however, a few reports are published on the cellulase production
using OP (Díaz et al., 2011) obtaining low enzymatic level of production. As for coffee husk, this
residue is mainly used as fertilizer or animal feed but the usage is limited due to the high amount of
organic material and phenolic compounds (antioxidants) (Belitz et al., 2009). For this reason, most of
the coffee husk is further stabilized by means of composting (Shemekite et al., 2014a).
CMCase ranged in all cases between 0.5-1.5 U g-1DM, which is in the lower range of the reported
(Table 1.3, section 1.5.1). BGase, in all cases was also produced in small amounts in comparison to
literature, with activity values below 0.2 U g-1DM (Table 1.3, section 1.5.1).
Chapter 4. Screening of wastes
67
Figure 4.6. FPase, CMCase and BGase activity values obtained by SSF at 37ºC using different substrates and compost as inoculum. The 1st sampling corresponds to the moment of maximum sOUR and the 2nd sampling was at the end of the fermentation. n.d. not determined.
CH+WC OP+WC OP+AH LP+AH OOC+WC
FPas
e [FP
U g-1
DM]
0
2
4
6
8
1st Sampling 2nd Sampling
CH+WC OP+WC OP+AH LP+AH OOC+WC
CMCa
se [U
g-1
DM]
0.0
0.5
1.0
1.5
2.0
1st Sampling 2nd Sampling
CH+WC OP+WC OP+AH LP+AH OOC+WC
BGas
e [U
g-1DM
]
0.0
0.2
0.4
0.6
0.8
1.0
1st Sampling 2nd Sampling
(n.d)
(n.d)
Chapter 4. Screening of wastes
68
The most interesting result was obtained by the experiments using OP and AH, where a considerable
BGase production of 0.111±0.003 U g-1DM was obtained in comparison with the results obtained
using the different substrates.
From these results, CH and OP were selected as the best substrates for further scale up assessment
4.2.2 SSF trials using orange peels as substrate
a) Operation in 10-L reactor
Process
This set of experiments were performed using the orange peels as a substrate and stabilized compost
as the inoculum. Also, due to the good results obtained in the previous section, almond hulls were
used as the bulking agent. Initial characterization of the raw materials and starting mixture for the 10
L reactors are presented in Table 4.3.
Table 4.3. Characterization of raw materials and initial mixture of 10-L OP reactor
Parameter OP Compost OP 10 L
Moisture (%) 82.3 35.3 62.0
Dry Matter (%) 17.7 64.7 38.0
Organic matter (%) 97.8 - 87.0
Total organic carbon (%) - - 27.0
C/N ratio - 15.0
pH 4.6 7.9 6.4
As seen in Table 4.3 , the starting pH of the process was 6.4. Shortly after the fermentation started,
there was a substantial drop in pH values due mainly to the organic acids released to the media, which
is expected in these processes. After that, a sustained increment of pH was obtained, reaching alkaline
values of 8.4 after 9 days of fermentation.
C/N ratio was 15 and 22 at the beginning and at the end of the fermentation, respectively. According
to literature, the optimum C/N ratio is around 20 or above (Puyuelo et al., 2011), depending on the
substrate. In this experiment, the initial C/N ratio is lower than the reported as optimum. The values of
carbon and nitrogen content observed in this experiment for this mixture are very different when
compared with the values obtained during the screening of the raw materials (Table 4.1, section 4.1).
Chapter 4. Screening of wastes
69
This can be due to the composition of the used compost or even to the heterogeneity of the mixture
itself., i.e some bulking agent traces may be present in the analyzed mixture.
In Figure 4.2a the full profiles of temperature and sOUR of process are presented. The SSF presented
an initial lag phase of nearly one day in the mesophilic range of temperature. After that, temperature
started to increase until 55-60°C between days 5 and 9 of the process. During the thermophilic stage,
highest sOUR was detected, reaching a maximum value of 5.6 mg g-1DM h-1 at day 7 of fermentation.
Figure 4.2. Profiles of a) Temperature and sOUR and b) enzymatic activity production for 10-L reactors using orange peels as substrate and compost as the inoculum.
Time [d]0 5 10 15 20 25 30
Tem
pera
ture
[°C
]
0
10
20
30
40
50
60
70
sOU
R [m
g O
2 g-1
DM
h-1
]
0
1
2
3
4
5
6
TemperaturesOUR
Time [d]
0 5 10 15 20 25 30
FPas
e
[FPU
g-1
DM
]
CM
Cas
e [U
g-1
DM
]B
Gas
e [U
g-1
DM
]
0
2
4
6
8
10
12
14
16
18
20
FPase CMCase BGase
a)
b)
Chapter 4. Screening of wastes
70
At the same time when maximum sOUR was found, maximum temperature was achieved, reaching a
value of 60ºC. In this moment, the reactor was supplied with a maximum airflow of 1000 mL min-1
(which is the maximum airflow possible using this configuration). In spite of that, oxygen content
drop below the recommended 10% to ensure full aerobic conditions (Ruggieri et al., 2009), reaching
values lower than 5%. This fact increases the possibility of the generation of anaerobic zones in the
solid matrix of the reactor. This, in a way, can affect the biodegradation process and the enzymatic
production.
After day 10 of operation, sOUR and temperature profiles presented a decreasing trend until reaching
mesophilic conditions and a low sOUR value, indicating a stabilization on biological activity. This
process achieved an overall COC of 285 g g-1DM.
Enzymatic activity production
In terms of enzymatic activity production, the complete profile is presented in Figure 4.2b. Regarding
cellulase activity production, it is possible to observe that FPase presented a high production, with a
peak of production during thermophilic stage that reached nearly 16 FPU g-1DM. This value is in
accordance to the reported range of cellulase activity production in SSF, which is 1-144 FPU g-1DM
(Dhillon et al., 2012b; El-Bakry et al., 2015; Mansour et al., 2016). However , it is important to
remark that for the measurement of cellulase activity there are several methods available with great
differences in the procedure for enzymatic determination and its consequents results (Percival Zhang
et al., 2006).
Regarding CMCase activity, the peak of production was obtained at day 3 of operation where the
system was already in fully thermophilic conditions. After that moment, the enzymatic production
decreased until almost negligible values in the 17th day of operation. The production of this enzyme
resulted in values several times lower than the obtained in other SSF using organic wastes as seen in
Table 1.3 section 1.5.1. CMCase reflects the endoglucanase activity which randomly bounds to any
section of the cellulose structure, reducing the polymerization degree of the cellulose chain. Also,
when working under the conditions presented in this SSF, it has to be taken into account that there are
several microorganisms working together and in competition for the same substrates. In addition,
considering that microorganisms excrete enzymatic compounds only when it is required, i.e when
soluble and easily metabolizable substrates are depleted; it is likely to assume that CMCase can have
a different profile (and peak) than the other cellulolytic components. In Figure 4.2b is clearly
observed than CMCase peak and complete profile is different that the observed for FPase and BGase.
Another reflection of these facts are the end products of cellulolytic action. Glucose and reducing
sugars content rapidly decreased during the process. At the beginning of the SSF reducing sugars and
Chapter 4. Screening of wastes
71
glucose content were near 0.3 g g-1DM and 0.065 g g-1DM respectively. After the first sampling at day
3 of fermentation, nearly 50 and 80% of the reducing sugars and glucose were consumed,
disappearing completely by day 9 in the case of glucose and day 17 in the case of reducing sugars.
Finally, for BGase activity, this also seems to be produced according to the requirements of the
microorganisms. It is observed that, in comparison with the other enzymes, there is no particular trend
in the production of BGase, obtaining two peaks of activity in days 5 and 9, with values of enzymatic
activity of roughly 0.4 U g-1DM.
In addition to the different trends among the cellulolytic components, it is clearly observed that
maximum production of FPase activity matches with the maximum sOUR and temperature.
According to Ghose, (1987) and Percival-Zhang et al., (2006), the FPase activity using filter paper as
substrate provides better information than other synthetic substrates of the potential hydrolytic effect
of cellulase on different real substrates. It is for this reason that most authors use this measurement as
a way to reflect a "total cellulase" measurement. In this sense, for the experiments performed in 10-L
reactors using coffee husk as the substrate only FPase was measured.
b) Operation in 50-L reactor.
Process
This experiment has two main differences with the 10L reactors: the first is the use of non-stabilized
digested compost as inoculum and the second is that in this case wood chips was used as bulking
agent. Duplicates of the reactors of the experiment were made (R1 and R2). Initial characterization is
presented in Table 4.4.
Table 4.4. Characterization of raw materials and initial mixture of 50-L OP reactor
Parameter OP Compost 50L OP 50L
Moisture (%) 82.3 29.9 58.1
Dry Matter (%) 17.7 70.1 41.9
Organic matter (%) 89.9
Bulk Density (g/L) 420.0
Air filled porosity (%) 74.0
C/N ratio 26.4
pH 4.6 6.6 5.9
Chapter 4. Screening of wastes
72
Figure 4.3 are presented the sOUR and temperature profiles of the pilot SSF (R1 and R2) using
orange peels as the substrate. In this figure it is possible to observe that, in general, all trends resulted
in very similar behavior. However slight differences can be detected in the temperature and sOUR
profiles that can be attributed to the intrinsic variability of the orange peel mixture, or even to the
differences in the design of the reactors itself. In this sense, thermophilic conditions were achieved in
the third day of process, although maximum temperature was not achieved until the twelfth day (R1:
67.4ºC; R2: 61.2ºC). The delay in temperature peaking, in comparison with the 10-L reactor, may be
due to different factors: the initial acidic pH of the mixture (5.9, Table 4.4) rose during the process
reaching a final value of 8.5; the lack of homogeneity of the initial mixture was overcome through
successive mixing when sampling, contributing to mass transfer between the different materials inside
the reactor; finally, a potential inhibition by the presence of some toxic component, such as limonene
that was stripped off the reactor by the continuous aeration. As expected, the highest sOUR values
(average of R1 and R2: 1.55±0.21 mgO2 g-1DM h-1, representing a coefficient of variation of 13%)
matched with the maximum temperature achieved (highest microbial activity). Also, the obtained
average COC of this fermentation was 243.9±19.7 mgO2 g-1DM, representing a variation coefficient
of 8%.
Enzymatic activity production
In Figure 4.4a is presented the full profile of cellulase activity and Figure 4.4b presents the reducing
sugars and glucose profiles during the fermentations. All values are expressed as the average value of
the two pilot reactors (R1 and R2).
FPase activity presented a continuous increase during the fermentation, peaking at 5.3±0.3 FPU g-
1DM during maximum sOUR period. Afterwards, the enzymatic production decreased until values
below 2 units of activity at the end of the fermentation. The slight disassociation of maximum
enzymatic activity with sOUR maximum can be attributed to a small differences among the reactors
performance.
CMCase activity profile is slightly different than the obtained for FPase, the maximum activity was
found at early stages of the process (at day 5 of fermentation) with a value of 3.531±0.003 U g-1DM.
When thermophilic stage started, CMCase action increased. It seems reasonable to hypothesized that
endocellulase production started earlier than FPase production, due to its role on depolymerization of
the cellulose structure and the lack of readily metabolizable sugars.
According to the results obtained at lab scale with OP as substrate, BGase activity was expected to
have a significant production and an accumulation in glucose content. Despite these findings, as seen
in Figure 4.4, BGase presented a small peak of production during the moment of maximum biological
Chapter 4. Screening of wastes
73
activity; however the magnitude of the obtained BGase activity surpassed the lab and bench scale
experiments, with a value of 0.48±0.07 U g-1DM. In terms of reducing sugars and glucose profile,
both reactors presented similar trends. A rapid reduction of both substrates was observed, which by
day 7 of operation were almost zero in both reactors. This behavior was practically identical as
obtained in 10L reactor.
Figure 4.3. Temperature and sOUR profiles for 50-L reactors R1 and R2, using orange peels as substrate and compost as the inoculum.
Time [d]0 5 10 15 20
Tem
pera
ture
[°C
]
0
10
20
30
40
50
60
70
80
sOU
R [m
g O
2 g-1
DM
h-1
]
0.0
0.5
1.0
1.5
2.0
Time [d]0 5 10 15 20
Tem
pera
ture
[°C
]
0
10
20
30
40
50
60
70
80
sOU
R [m
g O
2 g-1
DM
h-1
]
0.0
0.5
1.0
1.5
2.0
Temperature sOUR
R1
R2
Chapter 4. Screening of wastes
74
It has to be remarked the fact that even when working with large amounts of solid wastes in a pilot
scale and in duplicates, the obtained results are satisfactory in terms of the variation coefficient, due to
the high intrinsic variability of the process itself.
Figure 4.4. a) FPase, CMCase and BGase profiles and b) Reducing sugars and glucose profiles. Values obtained as an average between the two 50-L reactors R1 and R2, using orange peels as substrate and compost as the inoculum.
Time [d]0 5 10 15 20
FPas
e
[FPU
g-1
DM
]C
MC
ase
[U g
-1D
M]
BG
ase
[U g
-1D
M]
0
1
2
3
4
5
6FPase CMCase BGase
Time [d]
0 5 10 15 20
Red
ucin
g su
gars
[g g
-1D
M]
0.00
0.05
0.10
0.15
0.20
0.25
Glu
cose
[g g
-1D
M]
0.00
0.01
0.02
0.03
0.04
Reducing sugarsGlucose
a)
b)
Chapter 4. Screening of wastes
75
Table 4.5 shows a summary of the enzymatic production obtained in the experiments at the three
different scales. After a one-way ANOVA analysis it can be observed that there are significant
differences in the results obtained. It appears that when the SSF is performed in reactors of 10-L the
enzymatic production is higher than in other scales. It was expected that in the SSF performed in 0.5
L reactors the cellulase production was lower than at higher scales due to it was performed at fixed
temperature and using small amounts of substrate.
Table 4.5. Summary of the maximum filter paper activity (FPase) obtained in the three set of experiments carried out using range peels as substrate. Values of each enzymatic activity that do not share a letter are significantly different according to a one-way ANOVA analysis.
Experiments OP FPase
(FPU g-1 DM)
Reactor 0.5 L 1.6±0.5 (A)
Reactor 10 L 16.8±0.7 (B)
Reactor 50 L 5.3±0.3 (C)
It is possible that when working at the highest scale, some compacting issues appeared due to fungal
growth (Godoy et al., 2014) which could have led to a decrease in enzymatic production and the
generation of anaerobic/anoxic zones in the solid matrix. These contacting issues could have triggered
proliferation of anaerobic microorganisms and thus promoting the gaseous emissions of pollutants
such as methane of nitrous oxide. This point will be assessed in the following section.
Gaseous emissions
During orange peels (OP) solid state fermentation processes, measurements were made to determine
gaseous emissions. Volatile organic compounds (VOC), methane (CH4,), nitrous oxide (N2O) and
ammonia (NH3) were monitored. Table 4.6 presents the emission factors obtained from solid state
fermentation of orange peels. Also, the composition at family level of VOCs was analyzed and the
results are presented in Table 4.7. The emission factors and VOCs percentages presented the same
order of magnitude in both reactors
Chapter 4. Screening of wastes
76
Table 4.6. Emission factors obtained in 50-L pilot SSF using orange peels (OP) as substrate.
Kg VOC Mg -1OP Kg CH4 Mg -1OP Kg N2O Mg -1OP
R1 21.6 0.0120 0.0024
R2 13.4 0.0164 0.0066
Average±s.d 18±6 0.014±0.003 0.002±0.001
Table 4.7. Percentages of different VOCs families emitted during orange peels (OP) solid state fermentation in 50L reactors.
VOC family R1 (%) R2(%) Average ±s.d
Furans 0.1 0.2 0.2±0.1
Ester 0.9 2.9 1.9±1.4
Volatile Fatty Acids 0.3 1.2 0.8±0.6
Alcohols 4.0 6.1 5.1±1.5
Alyphatic hydrocarbons 0.2 1.2 0.7±0.7
Aldehydes 0.1 0.6 0.4±0.4
Ketones 1.2 2.0 1.6±0.6
Nitrogen containing compounds 0 0.1 0.10.1
Aromatic hydrocarbons 0.4 2.4 1.41.4
Terpenes 92.8 83.3 88.1±6.7
Sulphur containing compounds 0 0 0
Maximum methane production was obtained during thermophilic stage, during day 10 and 13 of
operation coinciding with sOUR maximum (Figure 4.5). Also, at the beginning of the fermentation (at
day 5), a small methane peak was observed. This can be due to the low initial pH in the reactors or to
a low air flow supply, which could have led to the generation of anaerobic zones. No ammonia was
emitted during this experiment, this was probably due to the low nitrogen content in the initial mixture
(1.4%).
Chapter 4. Screening of wastes
77
Nitrous oxide and VOCs also peaked during highest biological activity, in the thermophilic stage. The
main N2O emissions released during the process was found during the first week, as has been
previously reported for the aerobic degradation processes of other wastes (El Kader et al., 2007).
However, other authors related the inhibition of N2O production with thermophilic temperatures
(Fukumoto et al., 2003). In Figure 4.3 it can be observed that thermophilic temperatures were
achieved after the 10th of process pointing that N2O was emitted even at thermophilic temperatures.
N2O emission factors present the lowest values among the analyzed contaminants, as found in similar
biological processes (Maulini- Duran et al., 2013). In spite of that, from an environmental impact
point of view the contribution of N2O to global warming is significantly higher than CO2 or CH4.
Additionally, it should be highlighted that N2O presents the highest deviation obtained between
emission factors calculated for the two replicates. Along with methane emissions, VOCs also peaked
during the first days of operation. Probably, initial airflow supply generated stripping of the volatile
compounds across the solid matrix.
Regarding VOCs composition, all families emitted during the process are presented in Table 4.7.
Terpenes were the most emitted family of VOCs during the entire process, near 90% in both reactors.
The main compound emitted resulted to be limonene, which has been reported as antimicrobial
(Martin, 2010). The latter could have influenced in the biological activity during early stages of the
process and created a “lag phase” until day 5 of operation of the reactors. After that, an increase of air
flow supply took place, which could stimulate limonene releasing by stripping. Total emissions for
these compounds are presented in Table 4.8
Table 4.8. Total emission of selected individual volatile organic compounds
Reactors α- pinene
Kg Mg-1OP
β- pinene
Kg Mg-1OP
Limonene
Kg Mg-1OP
R1 0.017 0.029 0.43
R2 0.015 0.029 0.34
Average±s.d 0.016±0.001 0.029 0.39±0.06
Another important VOC released were α - and β -pinene, which are reported to be associated with
emissions generated of woody materials (Maulini-Duran et al., 2014). These compounds are also
reported to be related with the bulking agent selection (Büyüksönmez & Evans, 2007).
Chapter 4. Screening of wastes
78
4.2.3 SSF trials using coffee husk as substrate
a) Operation in 10-L reactor
Process
Figure 4.5 shows the profile of the process evolution of the solid state-fermentation of the coffee husk
using compost as the inoculum. It can be seen that the temperature profile followed a typical
composting pattern (Haug, 1993). After a brief lag phase, temperature rapidly began to increase up to
65°C, which can be attributed to the abundant and active indigenous microorganisms in the raw
materials. Thermophilic conditions were maintained for about one day and a half; after that,
temperature decreased to mesophilic values until the end of the experiment. Also, secondary
temperature increases were observed after homogenizing the reactor contents when sampling at days
3, 7, 10, 13 and 16.
During the thermophilic phase, it was observed a decrease of oxygen in the reactor until near 0% (
data not shown), which could have produced a loss of microbial activity, however after an adjustment
in the airflow, the oxygen content began to rise and the thermophilic conditions were restored, which
ensured metabolic activity (García et al., 2012). Also, it is important to mention that from the
thermophilic phase until the end of the operation of the reactor, pH was found around 9, which has
been found to be normal for lignocellulosic materials (García et al., 2012). Finally, maximum sOUR
obtained was 3.0 mgO2 g-1DM h-1 with a total COC of 1305 mgO2 g-1DM in 25 days of operation.
Enzymatic activity production
Figure 4.5 also shows the cellulase production profile along the SSF process. Filter paper activity
reached a maximum of 8.6 ±0.6 FPU g-1DM at the third day of operation. Cellulase production using
this system is in the reported range of cellulase production as seen in Table 1.3 section 1.5.1, although
most of these research has been performed in sterile conditions, mesophilic temperatures (30-35°C)
and variable moisture content (60-90%).
The highest cellulase activity was obtained at day 3 of fermentation, during the thermophilic phase as
reported in previous researches (Huang et al., 2010). This would be logical considering that is during
this stage that higher metabolic activity takes place (Barrena Gómez et al., 2005; Huang et al., 2010).
There are few researches on thermostable cellulases and most of them are related to modified
microorganisms at controlled and optimized operational conditions (Liu et al., 2011). Cellulase
activities obtained for thermostable enzymes ranged between 40-100 FPU g-1DM (Liu et al., 2011).
A decrease of cellulase production was observed after the third day peak, which is probably due to the
irreversible adsorption of the enzyme onto cellulose or lignin or even to the depletion of nutrients
Chapter 4. Screening of wastes
79
(Gusakov & Sinitsyn, 1992; Olofsson et al., 2008; Xiao et al., 2004). With the aim to achieve higher
amount of cellulase different alternatives have been evaluated to enhance cellulase production such as
alkali or acid pretreatment or inductors addition (Bansal et al., 2012; Dhillon et al., 2012a).
Figure 4.4. Filter paper activity (FPase), temperature and sOUR profiles for 10-L reactors using coffee husk as substrate and compost as the inoculum.
b) Operation in 50-L reactor
Process
Initial characterization of the raw materials and initial mixture for the experiments in 50L reactors of
coffee husk are presented in Table 4.9.
In Figure 4.5 temperature and sOUR profiles are presented for the two pilot SSF using coffee husk as
substrate (R1 and R2). During the development of these experiments, an aeration problem arose in
R2, which lead to an irregular air flow and malfunction of the airflow controller. Is it for this reason
that sOUR profile in R2 (Figure 4.5) presents a long period of noise since day 2 of operation. In spite
of these operational difficulties, both fermentations showed a clearly similar trend in both monitored
parameters.
Time [d]
0 5 10 15 20 25
Tem
pera
ture
[°C
]
0
10
20
30
40
50
60
70
80
sOU
R [m
g O
2 g-1
DM
h-1
]
0.0
0.5
1.0
1.5
2.0
2.5
FPas
e [F
PU g
-1D
M]
0
2
4
6
8
10
Temperature sOUR FPase
Chapter 4. Screening of wastes
80
Table 4.9. Characterization of raw materials and initial mixture of 50L CH reactor.
Parameter CH Compost 50L CH 50L
Moisture (%) 59.8 36.9 59.5
Dry Matter (%) 40.1 63.1 40.5
Organic matter (%) 90.2 90.0
Bulk Density (g/L) 238.1 358.0
Air filled porosity (%) 78.9 77.2
C/N ratio 22.8 13.5
pH 6.3 7.5 7.2
Maximum temperature was achieved in day 2 of operation, reaching values of 66.5 and 75.3 in R1
and R2 respectively. Fully thermophilic stage lasted nearly 3 days in both reactors, after that,
mesophilic conditions were reached until the end of the fermentation. It is possible that temperatures
as high as the obtained in R2 could negatively affect the process, in terms of reducing the microbial
diversity and activity which will directly affect the enzymatic activity production (Arora et al., 2015;
Ekinci et al., 2004).
Regarding sOUR, maximum values were obtained also during thermophilic stage, in day 2 of
operation reaching values of 2.5 and 3.3 mg O2 g-1DM h-1for R1 and R2 respectively. Similarly to the
obtained in the pilot experiments with orange peels, the average results did not present a big
dispersion. In this sense, the average sOUR of the process is 2.9 ±0.6 mg O2 g-1DM h-1 with a
variation coefficient of 20%. Additionally, the average cumulative oxygen consumption between R1
and R2 was 171± 26 mg O2 g-1DM, which a variation coefficient of 15%.
During this fermentation C/N ratio was also monitored. The initial C/N value was 13.5, which is a low
value in comparison with another substrates and far from the recommended 25 (Jiang et al., 2011).
The entire process achieved a total C/N ratio reduction of 13.5% in 15 days of operation. This
reduction may seem small, however, authors working in co composting with coffee husk obtained
nearly 14.2% reduction in 32 days of operation and with an initial C/N ratio of 17.5 (Shemekite et al.,
2014b). Low C/N ratio in the present fermentations, are mainly due to unusually low carbon content
in the initial mixture. C content was roughly 40% . This is not comparable with obtained 64% in
previous fermentations (C/N ratio:24.8) (Abraham et al., 2013). The intrinsic variability of coffee
Chapter 4. Screening of wastes
81
husk and compost used in these fermentations and of course a lack of homogeneity in the sampling
could have affected these measurements.
Figure 4.5. Temperature and sOUR profiles for 50-L reactors R1 and R2, using coffee husk as substrate and compost as the inoculum
Time [d]0 5 10 15
Tem
pera
ture
[°C
]
0
10
20
30
40
50
60
70
80
sOU
R [m
gO2 g-1
DM
h-1
]
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
Time [d]0 5 10 15
Tem
pera
ture
[°C
]
0
10
20
30
40
50
60
70
80
sOU
R [m
gO2 g-1
DM
h-1
]
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
Temperature sOUR
R2
R1
Chapter 4. Screening of wastes
82
Enzymatic activity production
In Figure 4.6a shows the average cellulases production profile obtained in the 50 L SSF. Also Figure
4.6b presents the reducing sugars and glucose content during the fermentation in the reactor.
Enzymatic profile of the group of cellulases show clearly a maximum during the time of maximum
sOUR and temperature, which is in accordance to the previous experiments carried out in 10 L and
0.5 L reactors. Regarding FPase activity, it reached average values of nearly 9.5±0.7 FPU g-1DM.
These values are similar to the obtained in the 10 L SSF which can, in a way, imply that the scale
factor is not of great importance when working under the specified conditions ans usinf this specific
substrate. In Table 4.10 is summarized the cellulase production at the three different scales. The
ANOVA analysis confirms that there is no significant differences between the results obtained at 10
or 50 L.
Table 4.10. Summary of the maximum filter paper activity (FPase) obtained in the three set of experiments carried out using coffee husk as substrate. Values of each enzymatic activity that do not share a letter are significantly different according to a one-way ANOVA analysis.
Experiments CH FPase
(FPU g-1 DM)
Reactor 0.5 L 6.4±0.7 (A)
Reactor 10 L 8.6 ±0.6 (B)
Reactor 50 L 9.5±0.7 (B)
The lack of an scale impact on enzymatic production is in accordance to the obtained by Astolfi et al.,
(2011). They assessed the scale impact on enzymatic production using on one hand 5g and on the
other hand, 2 Kg of substrate, resulting in statistically similar activity values.
As mentioned before, several differences in enzymatic production among the replicates could have
been generated due to aeration difficulties, specifically due to air supply fluctuations and the
generation of anaerobic/anoxic conditions (Rodriguez-Fernandez et al., 2012). In the same context,
Rodriguez- Fernandez et al., (2012) has also stated that proper airflow supply is a key factor when a
scale-up is considered. In spite of that, the enzymatic production in the replicates are very similar. As
for FPase, the variation coefficient obtained as between nearly 4 and 40%, where the maximum
dispersion was found at the moment of maximum fluctuation in airflow which could have lead to the
mentioned differences. However, it is important to mention that obtained FPase values at pilot-scale
Chapter 4. Screening of wastes
83
are found in the upper bound of the reported activities (Table 1.3) obtained in sterile conditions,
mesophilic temperatures (30-35°C) and variable moisture content (60-90%).
Figure 4.6. a) FPase, CMCase and BGase profiles and b) Reducing sugars and glucose profiles. Values obtained as an average between the two 50-L reactors R1 and R2, using coffee husk as substrate and compost as the inoculum.
Time [d]0 5 10 15
FPas
e
[FPU
g-1
DM
]C
MC
ase
[U g
-1D
M]
BG
ase
[U g
-1D
M]
0
2
4
6
8
10
12FPase CMCase BGase
Time [d]
0 5 10 15
Red
ucin
g su
gars
[g
g-1D
M]
0.000
0.005
0.010
0.015
0.020
0.025
0.030
0.035
Glu
cose
[g g
-1D
M]
0.0000
0.0001
0.0002
0.0003
0.0004
0.0005
Reducing sugarsGlucose
a)
b)
Chapter 4. Screening of wastes
84
BGase production profile is the process where major differences were founds in the replicates. The
coefficients of variation in all samplings were above 40%, which suppose a high dispersion of the
production between the processes. In addition to the above mentioned airflow difficulties, it is
possible that the differences could have been due to performance of the reactors or even the sampling
for enzymatic measurements were lacking of homogeneity. Another main factor to consider in
cellulase activity production is temperature. Cellulases optimum temperature ranges between 50-
55°C, for cellulase of fungal origin. High temperatures obtained in R2 could influence BGase activity,
in comparison with R1.
Also it is important to consider the microorganisms who are able to produce cellulases and that are
present in both reactors. Some authors have reported that even knowing that bacteria and fungi are
able to use cellulose as primary carbon source, bacteria are incapable of degrading crystalline
cellulose since their cellulase systems are incomplete. This is a different situation for fungi, that are
able to degrade completely this structure (Brijwani et al., 2011).
In this context, it also has been reported by Shatemike et al., (2014) that fungi are not able to thrive
under rough environmental conditions such as the early stage of this SSF or even in a composting
process, it is for this reason that fungi appears at latter stage of this process and cellulases activity
appears at day 32 of composting. However other authors (Lever et al., 2010), has proven to increase
cellulase activity and fungi growth in a SSF process by using fermented solids as inoculum. The
latter, supports the fact that using compost as an initial mixed inoculum provides different
microorganisms (among them fungi and yeasts) that are able to quickly colonize the reactor. In
addition, coffee husk provides an ideal environment for fungi to thrive
Figure 4.8b shows the average reducing sugars and glucose profile of the pilot SSF. The glucose
profile, shows a rapid consumption at day 2 of operation which left almost negligible amounts of
glucose detected in the reactor at day 5. During this stage the reducing sugar content remained almost
unchanged until day 7 of fermentation where reducing sugars started to be consumed and an slight
glucose accumulation was observed. In contrast of the findings on the OP experiments, the reducing
sugars content did not disappeared of the solid matrix, but only decreased in a 33% . This can be a
reflection of CMCase or FPase action which could have release hydrolyzates of cellulose (or
hemicellulose) with reducing ends exposed to the media.
Gaseous emissions
As it is possible to observe, all emissions presented maximum production during thermophilic stage.
In the case of VOCs trends resulted similar in both reactors and obtained similar emission factors, as
shown in Table 4.11, with values of 0.14 and 0.12 Kg Mg-1CH.
Chapter 4. Screening of wastes
85
Nitrous acid production, presented high production at the first days of fermentation, during
thermophilic stage; after that, nitrous acid was not significantly produced. It has been reported that a
temperature, nitrogen content and aeration rates influences nitrous oxide emissions and found to be
produced mainly during early stages of composting processes (He et al., 2001). Also alkaline pH
values enhances nitrous acid production, and considering that since day 2 of operation pH ranges
between 8.5-9.2, it was expected an increase of nitrous acid. In this case in particular, emission factor
obtained was 0,04 and 0,017 Kg Mg-1CH for R1 and R2. It is possible that fluctuations in aeration
rates in R1, could have influenced higher emissions in all evaluated compounds.
On the other hand, methane production presented different trends in both reactors. R1, presented a fast
methane peak during thermophilic stage, however after that, no significant methane production was
observed, even when aeration rate was low (400 mL min-1) during the last stage of the fermentation.
In R2, a higher methane production was achieved. At the beginning of the fermentation high rate of
methane production was observed until day 5 of operation. However, instead of stop producing
methane as in R1, it kept increasing, although at lower rate. Emission factors in R1 and R2 were
0.0007 and 0.0019 Kg Mg-1CH respectively.
It is possible that, considering methane emissions, anaerobic or anoxic zones could have been
generated in R2. This could be confirmed by low oxygen level in the reactor during thermophilic
stage, lower oxygen consumption and maybe for insufficient aeration during this period.
Regarding VOCs composition, all families emitted during the process are presented in Table 4.12.
Table 4.11. Emission factors obtained in coffee husk (CH) solid state fermentation in 50L reactors.
Reactors VOC
Kg Mg-1 CH
CH4
Kg Mg-1 CH
N2O
Kg Mg-1 CH
R1 0.14 0.0007 0.04
R2 0.12 0.0019 0.02
Average 0.13±0.014 0.0013±0.0008 0.02±0.02
As in orange peel experiments, terpenes are the main family emitted during coffee husk solid state
fermentation, with 61.6 and 66.6% in R1 and R2. This can be expected due to coffee husk
composition includes terpenes as a main responsible for flavour and aromatic in this residue (Ray &
Ward, 2008). In the first stages of the fermentation, a variety of nearly 10 different terpenes were
Chapter 4. Screening of wastes
86
emitted, however after thermophilic stage three compounds were mainly produced: limonene,
camphene (an isomeric form of limonene) and α -β- pinene. In Table 4.13 total emission of selected
compounds are presented. All these components are related to flavour properties in coffee, however,
pinene forms can be also related to bulking agent and its woody origin.
Second most important VOC emitted were ketones, with an average of 15.8% in both reactors. This
was expected, considering coffee husk volatile fraction composition, which includes nearly 20% of
ketones (van Dam & Harmsen, 2010).
Table 4.12. Percentages of different VOCs families emitted in coffee husk (CH) solid state fermentation in 50L reactors.
VOC family R1 (%) R2(%) Average
Furans 0.52 0.16 0.34±0.25
Ester 0.00 0.00 0.0±0.0
Volatile Fatty Acids 4.21 0.00 2.10±2.98
Alcohols 0.74 2.55 1.65±1.28
Alyphatic hydrocarbons 5.69 7.28 6.49±1.12
Aldehydes 4.85 0.00 2.43±3.43
Ketones 15.70 15.95 15.28±0.18
Nitrogen containing compounds 1.44 2.00 1.72±0.39
Aromatic hydrocarbons 5.26 4.95 5.10±0.22
Terpenes 61.6 66.6 64.1±3.54
Sulphur containing compounds 0.01 0.00 0.005±0.007
Halogenated compounds 0.00 0.51 0.26±0.36
Chapter 4. Screening of wastes
87
Table 4.13. Total emission of selected individual volatile organic compounds.
Reactors α- pinene
Kg Mg-1CH
β- pinene
Kg Mg-1CH
Limonene
Kg Mg-1CH
R1 0.0016 0.0002 0.0011
R2 0.0010 0.0001 0.0011
Average 0.0013±0.0004 0.0002±0.0001 0.0011
4.3 Conclusion
It was possible to determine at lab scale that orange peels and coffee husk were the most promising
raw materials to use as substrate for solid state fermentation for cellulase production. When the scale
up was assessed in a 10 L reactor, a decrease of nearly 50% was observed in cellulase production.
Finally when the pilot scale experiments were performed, the enzymatic production in comparison
with the 10 L SSF was sustained, achieving a maximum value of 5.3±0.3 and 9.5±0.7 FPU g-1DM for
orange peels and coffee husk, respectively. The emission factors of both orange peels and coffee husk
are significantly lower than the observed in similar processes. In both pilot fermentation the main
emitted gases were the VOCs which is probably due to the high carbonaceous material and phenolic
compounds present in both raw materials. Among all VOCs, the terpenes were the family with the
highest emission factor, followed by α and β- pinene, which are compounds related to lignocellulosic
material degradation. The success of both fermentations, especially using coffee husk as the substrate
in terms of enzymatic production and environmental impact makes this substrate selected for further
research and optimization.
References
Abraham, J., Gea, T., Sánchez, A. 2013. Potential of the solid-state fermentation of soy fibre residues by native microbial populations for bench-scale alkaline protease production. Biochemical Engineering Journal, 74, 15-19.
Arora, R., Behera, S., Kumar, S. 2015. Bioprospecting thermophilic/thermotolerant microbes for production of lignocellulosic ethanol: A future perspective. Renewable and Sustainable Energy Reviews, 51, 699-717.
Astolfi, V., Joris, J., Verlindo, R., Oliveira, J., Maugeri, F., Mazutti, M., de Oliveira, D., Treichel, H. 2011. Operation of a fixed-bed bioreactor in batch and fed-batch modes for production of inulinase by solid-state fermentation. Biochemical Engineering Journal, 58–59, 39-49.
Bansal, N., Tewari, R., Soni, R., Soni, S. 2012. Production of cellulases from Aspergillus niger NS-2 in solid state fermentation on agricultural and kitchen waste residues. Waste Management, 32(7), 1341-1346.
Chapter 4. Screening of wastes
88
Barrena Gómez, R., Vázquez Lima, F., Gordillo Bolasell, M., Gea, T., Sánchez Ferrer, A. 2005. Respirometric assays at fixed and process temperatures to monitor composting process. Bioresource Technology, 96(10), 1153-1159.
Belitz, H., Grosch, W., Schieberle, P. 2009. Food Chemistry. 4 ed, Springer-Verlag Berlin Heidelberg. Berlin, Germany, pp. 1070.
Büyüksönmez, F., Evans, J. 2007. Biogenic Emissions from Green Waste and Comparison to the Emissions Resulting from Composting Part II: Volatile Organic Compounds (VOCs). Compost Science & Utilization, 15(3), 191-199.
Colon, J., Cadena, E., Pognani, M., Barrena, R., Sanchez, A., Font, X., Artola, A. 2012. Determination of the energy and environmental burdens associated with the biological treatment of source-separated Municipal Solid Wastes. Energy & Environmental Science, 5(2), 5731-5741.
Dhillon, G., Brar, S., Kaur, S., Metahni, S., M'Hamdi, N. 2012a. Lactoserum as a moistening medium and crude inducer for fungal cellulase and hemicellulase induction through solid-state fermentation of apple pomace. Biomass and Bioenergy, 41, 165-174.
Dhillon, G., Kaur, S., Brar, S., Verma, M. 2012b. Potential of apple pomace as a solid substrate for fungal cellulase and hemicellulase bioproduction through solid-state fermentation. Industrial Crops and Products, 38(1), 6-13.
Díaz, A., Bolívar, J., de Ory, I., Caro, I., Blandino, A. 2011. Applicability of enzymatic extracts obtained by solid state fermentation on grape pomace and orange peels mixtures in must clarification. LWT - Food Science and Technology, 44(4), 840-846.
Ekinci, K., Keener, H., Michel, F., Elwell, D. 2004. Modeling Composting Rate as a Function of Temperature and Initial Moisture Content. Compost Science & Utilization, 12(4), 356-364.
El-Bakry, M., Abraham, J., Cerda, A., Barrena, R., Ponsá, S., Gea, T., Sánchez, A. 2015. From Wastes to High Value Added Products: Novel Aspects of SSF in the Production of Enzymes. Critical Reviews in Environmental Science and Technology, 45(18), 1999-2042.
El Kader, N., Robin, P., Paillat, J., Leterme, P. 2007. Turning, compacting and the addition of water as factors affecting gaseous emissions in farm manure composting. Bioresour Technol, 98(14), 2619-28.
Fukumoto, Y., Osada, T., Hanajima, D., Haga, K. 2003. Patterns and quantities of NH(3), N(2)O and CH(4) emissions during swine manure composting without forced aeration--effect of compost pile scale. Bioresour Technol, 89(2), 109-14.
García, A., Maulini, C., Torrente, J., Sánchez, A., Barrena, R., Font, X. 2012. Biological treatment of the organic fibre from the autoclaving of municipal solid wastes; preliminary results. Biosystems Engineering, 112(4), 335-343.
Ghose, T. 1987. Measurement of cellulase activities. Pure and Appied Chemistry, 59(2), 257-268. Godoy, M., Roussos, S., Freire, D. 2014. Respirometric analysis of Penicillium simplicissimum
growth in solid-state fermentation using Jatropha cake as culture medium. BMC Proceedings, 8(Suppl 4), P220-P220.
Grohmann, K., Baldwin, E. 1992. Hydrolysis of orange peel with pectinase and cellulase enzymes. Biotechnology Letters, 14(12), 1169-1174.
Gusakov, A., Sinitsyn, A. 1992. A theoretical analysis of cellulase product inhibition: Effect of cellulase binding constant, enzyme/substrate ratio, and β-glucosidase activity on the inhibition pattern. Biotechnology and Bioengineering, 40(6), 663-671.
Haug, R. 1993. The Practical Handbook of Compost Engineering, Lewis Publishers. Boca Raton, Florida.
He, Y., Inamori, Y., Mizuochi, M., Kong, H., Iwami, N., Sun, T. 2001. Nitrous Oxide Emissions from Aerated Composting of Organic Waste. Environmental Science & Technology, 35(11), 2347-2351.
Huang, D., Zeng, G., Feng, C., Hu, S., Lai, C., Zhao, M., Su, F., Tang, L., Liu, H. 2010. Changes of microbial population structure related to lignin degradation during lignocellulosic waste composting. Bioresource Technology, 101(11), 4062-4067.
Chapter 4. Screening of wastes
89
Jiang, T., Schuchardt, F., Li, G., Guo, R., Zhao, Y. 2011. Effect of C/N ratio, aeration rate and moisture content on ammonia and greenhouse gas emission during the composting. Journal of Environmental Sciences, 23(10), 1754-1760.
Li, P., Xia, J., Nie, Z., Shan, Y. 2016. Pectic oligosaccharides hydrolyzed from orange peel by fungal multi-enzyme complexes and their prebiotic and antibacterial potentials. LWT - Food Science and Technology, 69, 203-210.
Liu, D., Zhang, R., Yang, X., Wu, H., Xu, D., Tang, Z., Shen, Q. 2011. Thermostable cellulase production of Aspergillus fumigatus Z5 under solid-state fermentation and its application in degradation of agricultural wastes. International Biodeterioration & Biodegradation, 65(5), 717-725.
Mansour, A., Arnaud, T., Lu-Chau, T., Fdz-Polanco, M., Moreira, M., Rivero, J. 2016. Review of solid state fermentation for lignocellulolytic enzyme production: challenges for environmental applications. Reviews in Environmental Science and Bio/Technology, 15(1), 31-46.
Maulini-Duran, C., Artola, A., Font, X., Sánchez, A. 2014. Gaseous emissions in municipal wastes composting: Effect of the bulking agent. Bioresource Technology, 172, 260-268.
Moosavi Dolatabadi, K., Dehghan, G., Hosseini, S., Jahanban Esfahlan, A. 2015. Effect of five year storage on total phenolic content and antioxidant capacity of almond (Amygdalus communis L.) hull and shell from different genotypes. Avicenna Journal of Phytomedicine, 5(1), 26-33.
Olofsson, K., Bertilsson, M., Lidén, G. 2008. A short review on SSF – an interesting process option for ethanol production from lignocellulosic feedstocks. Biotechnology for Biofuels, 1(1), 1-14.
Percival-Zhang, Y., Himmel, M., Mielenz, J. 2006. Outlook for cellulase improvement: Screening and selection strategies. Biotechnology Advances, 24(5), 452-481.
Puyuelo, B., Ponsá, S., Gea, T., Sánchez, A. 2011. Determining C/N ratios for typical organic wastes using biodegradable fractions. Chemosphere, 85(4), 653-659.
Ray, R., Ward, P. 2008. Microbial Biotechnology in Horticulture, Vol. 3, CRC Press, Taylor & Francis Group. Florida, USA.
Rodriguez-Fernandez, D., Rodriguez-Leon, J., de Carvalho, J., Karp, S., Sturm, W., Parada, J., Soccol, C. 2012. Influence of airflow intensity on phytase production by solid-state fermentation. Bioresour Technol, 118, 603-6.
Ruggieri, L., Gea, T., Artola, A., Sánchez, A. 2009. Air filled porosity measurements by air pycnometry in the composting process: A review and a correlation analysis. Bioresource Technology, 100(10), 2655-2666.
Shemekite, F., Gómez-Brandón, M., Franke-Whittle, I., Praehauser, B., Insam, H., Assefa, F. 2014a. Coffee husk composting: An investigation of the process using molecular and non-molecular tools. Waste Management (New York, N.y.), 34(3), 642-652.
Shemekite, F., Gómez-Brandón, M., Franke-Whittle, I., Praehauser, B., Insam, H., Assefa, F. 2014b. Coffee husk composting: An investigation of the process using molecular and non-molecular tools. Waste Management, 34(3), 642-652.
Sundberg, C., Yu, D., Franke-Whittle, I., Kauppi, S., Smårs, S., Insam, H., Romantschuk, M., Jönsson, H. 2013. Effects of pH and microbial composition on odour in food waste composting. Waste Management (New York, N.y.), 33(1), 204-211.
van Dam, J., Harmsen, P. 2010. Coffee residues utilization, Wageningen UR Food and Biobased research. Wageningen, The Neatherlands., pp. 75.
Xiao, Z., Zhang, X., Gregg, D., Saddler, J. 2004. Effects of sugar inhibition on cellulases and beta-glucosidase during enzymatic hydrolysis of softwood substrates. Applied Biochemistry and Biotechnology, 113-116, 1115-26.
Chapter 4. Screening of wastes
90
Chapter 5. Towards a competitive SSF
91
CHAPTER 5. TOWARDS A COMPETITIVE
SOLID-STATE FERMENTATION FOR ENZYME
PRODUCTION
Part of this chapter has been submitted for publication in Science of the Total Environment. "Towards
a competitive solid state fermentation: cellulases production from coffee husk by sequential batch
operation and role of microbial diversity".
Cerda, A., Vargas-García, M.C., Gea, T., Sánchez, A.
Chapter 5. Towards a competitive SSF
92
Chapter 5. Towards a competitive SSF
93
Summary
In this chapter are presented operational strategies for the development of a continuous SSF for
cellulase production. As mentioned in Chapter 1, one of the main constrains of SSF is the difficulty to
work in a continuous configuration. In this sense, in section 5.1 is presented the development of a
sequential batch operation for cellulase using different substrate exchange ratios.
5.1 Towards a competitive solid state fermentation: cellulases production from coffee husk by
sequential batch operation and role of microbial diversity.
5.1.1 Materials
Coffee husk (CH) was kindly provided by Marcilla S.A (Mollet del Vallés, Barcelona, Spain) and
compost (C, from source-selected organic fraction of municipal solid waste) was obtained from the
municipal solid waste treatment plant Ecoparc de Montcada (Montcada, Barcelona, Spain) as
explained in chapter 3 section 3.1.1. All materials were stored frozen (-18ºC) until use. The full
characterization of both raw materials and the initial and final composition of reactors is presented in
Table 5.1.
5.1.2 Experimental procedure
a) Selection of inoculum size
Preliminary SSF experiments were performed in order to assess the inoculum size and the time for
maximum cellulase production. Non sterile coffee husk was mixed with compost as inoculum in 0,
10, 20, 50 and 100% (w/w) ratio. Wood chips were added in a 1:1 (v/v) ratio as bulking agent.
Fermentation was carried out using the system described in chapter 3 section 3.2.1 SSF were
performed in triplicates for 4 days at 37ºC using 90 g of the mixture CH and C using continuous
aeration of 20 mL min-1 and oxygen monitoring.
b) Sequential batch operation for cellulase production
Fermentations were performed in 4.5L air-tight packed-bed reactors (adapted Dewar® vessels),
thermally isolated to work under near-to-adiabatic conditions (Dewar® vessels provide excellent
thermal isolation, minor heat exchange with the surroundings take place through the airflow (in and
out) and partially through the lid). Experimental set-up is presented in Chapter 3, section 3.2.2 (Figure
3.3).
Chapter 5. Towards a competitive SSF
94
A self-made acquisition and control system was used based on Arduino® and self-made software. Air
was continuously supplied to the reactors by means of a mass flow controller (Bronkhorst, Spain).
Airflow was automatically adjusted by a feedback controller (aeration range 15.8 – 31.6 L kg-1 DM h-
1, set point 11.5% oxygen in exhaust gas) as described in chapter 3 section 3.4.1. Airflow, temperature
and oxygen content were continuously monitored.
The initial mixture for the first batch in both strategies contained 90% (wb) of fresh coffee husk as
substrate and 10%(wb) of compost as mixed inoculum. Compost was added in order to provide active
biomass and mainly for the potential incorporation of diverse communities able to degrade
lignocellulosic materials, as reported in recent studies (Lopez-Gonzalez et al., 2014). Wood chips
were added as bulking agent in a ratio of 1:1 (v/v) in order to provide enough porosity to promote
proper oxygen transfer (Ruggieri et al., 2009). Fermentations were performed with a total weight of
1.2 kg per batch.
Table 5.1. Characterization of coffee husk (CH), compost (C), initial mixtures and final fermented
solids for SB50 and SB90. wb: wet basis; db:drybasis
Parameter Coffee Husk Compost
SB50
initial
SB90
initial
SB50
Final
SB90
final
Moisture
(% wb)
60.8 ± 0.4 34.9 ± 0.3 61.2 ±1.1 62.0 ± 1.3 55.0 ± 1.1 55.0 ± 3.1
pH 6.51 ± 0.01 7.22 ± 0.01 6.45 ± 0.01 6.43 ± 0.01 9.11 ± 0.01 9.07 ± 0.01
Cellulose
(%, db)
25.7 ± 0.2 9.8 ± 1.2 23.4 ± 0.8 27.9 ± 0.4 28.6 ± 0.5 23.5 ± 0.4
Hemicellulose
(%,db)
14.6 ± 0.1 10.2 ± 0.1 13.2 ± 0.5 13.3 ± 0.3 13.2 ± 0.5 13.3 ± 0.5
Lignin
(%, db)
17.6 ± 0.5 13.6 ± 1.2 20.4 ± 0.6 21.8 ± 0.6 16.5 ± 0.7 18.3 ± 0.6
Maximum cellulases activity was achieved in 48h and thus the solid retention time (SRT) was
established in 48h. Sequential batch operation was performed in two different configurations:
SB90: 90% of the wet fermented solids were removed from the reactor at the end of the fermentation
(every 48h) and used as product for final analysis. The remaining 10% of the fermented solids acted
as inoculum to start a new batch, using 90% fresh coffee husk. Six sequential batches (12 days) were
performed to allow the microbial community to develop and to assess the reproducibility of the
cellulases production.
Chapter 5. Towards a competitive SSF
95
SB50: 50% of the wet fermented solids were removed from the reactor every 24h and used as product
for further analysis. In this case, the remaining 50% of the fermented solid was used as inoculum to
start a new batch, with the addition of 50% of fresh coffee husk. SB50 operated for 18 days.
Both strategies performed until steady operation resulting in 12 days for SB90 and 18 days for SB50.
During these processes, a continuous on-line monitoring of temperature and sOUR was carried out.
Sampling was always performed after a complete manual homogenization of the fermented solids to
obtain a full representative sample and prior to feeding the reactor with fresh substrate. From these
solid samples, measurements of elemental composition (C,H,O and N), fiber content (cellulose, lignin
and hemicellulose), microbial diversity, enzymatic activity and other routine methods (Chapter 3,
section 3.3) were determined.
c) Cellulase production profile using specialized inoculum
The final fermented solids obtained at the end of both sequential batch operations SB90 and SB50
were used to inoculate fresh coffee husk in a 10% (w/w) ratio for a new single batch to analyze
whether the selected biomass modified the process dynamics and the cellulase production profile.
Experiments were performed in the 4.5L bioreactors with a total mass of 1.2 kg and samples were
collected after complete homogenization of the reactors at 0, 24, 36, 40 and 60h.
5.1.3 Specific measurements for biodiversity
Identification of the microbial population was performed in the raw materials and in the final samples
from both SB50 and SB90 operational strategies using next generation sequencing. The aim of this
analysis was to determine the potential variation on biodiversity and to characterize the specialized
biomass obtained during sequential batch operations.
Total DNA was extracted and purified using PowerSoil™ DNA Isolation Kit (MoBio Laboratories,
USA) according to provider’s specifications. DNA samples were checked for concentration and
quality using the NanoDrop ND-1000 Spectrophotometer (NanoDrop Technologies, Wilmington,
Delaware; USA) (Lopez-Gonzalez et al., 2014).
Bacterial 16S rRNA hypervariable regions V3-V4 and Fungal ITS1-ITS3 were targeted. Later
sequencing of the extracted DNA and bioinformatics were performed on MiSeq platform by Life
Sequencing S.A (Valencia, Spain).
Shannon-Wiener (H´) and qualitative (IS) and quantitative (ISquant) Sorensen´s-Dice biodiversity
indices were estimated according to the following equations:
𝐻´ = − 𝑝!𝑙𝑜𝑔!𝑝!
where pi = ni/N, ni= relative abundance of the ith species and N =Σn, and
Chapter 5. Towards a competitive SSF
96
𝐼! = !!!!!
; 𝐼!"#$%& = !!"!"!!"
where a = number of species in sample 1, b = number of species in sample 2, c = number of species
shared by the two samples, pN = sum of the lower of the two abundances recorded for species found
in the two samples, aN = number of individuals in sample 1, bN = number of individuals in sample.
5.1.4 Results
a) Selection of inoculum size
Figure 5.1a shows the profiles of the cellulase activity obtained when using different inoculum sizes.
Maximum cellulase activity was found in the first day of operation in all cases. It can be noticed that
the same trend is observed in all experiments; a rapid increase in cellulase activity in day one
followed by a gradual drop until the end of the fermentation. The maximum cellulase activity was
statistically similar for samples with 10 and 20% added compost, reaching 7.0±0.3 and 7.5±0.5 FPU
g-1 DM respectively. After day one, cellulase production decreased achieving values around 0.18 FPU
g-1DM by the end of the fermentation. Adding 50% compost dramatically reduced cellulases
production until almost negligible values below values obtained with only CH or only C.
In previous studies, the maximum enzymatic activity of amylases or proteases was not related to the
maximum biological activity measured as sOUR in similar systems (El-Bakry et al., 2016, Cerda et
al., 2016). However, for cellulases the opposite pattern was observed. The maximum cellulase
production was found on the peak of biological activity, i.e. during maximum sOUR for all the
assessed inoculum ratios (data not shown). As an example, Figure 5.1b shows a full sOUR profile for
the sample using 10% of compost as inoculum. In this Figure, it can be observed that the maximum
cellulase activity was detected during the most active biooxidative stage of the process, as other
authors have also stated (Jurado et al., 2014). Similar results were observed in batch experiments
performed in 4.5 L bioreactors where a process time of 48h was fixed for maximum sOUR and
cellulase production.
Chapter 5. Towards a competitive SSF
97
Figure 5.1. SSF of coffee husk using compost as inoculum at lab scale and 37ºC during 4 days. a)
Profile of cellulase production using different inoculum sizes and b) profiles of sOUR and cellulase
production using 10% of inoculum (CH: coffee husk; C:compost).
Time [d]
0 1 2 3 4
sOU
R [m
gO2
g-1D
M h
-1]
0.0
0.5
1.0
1.5
2.0
Cel
lula
se a
ctiv
ity [F
PU g
-1D
M]
0
2
4
6
8
Time [d]0 1 2 3 4
Cel
lula
se a
ctiv
ity [F
PU g
-1D
M]
0
1
2
3
4
5
6
7
8
a)
b)
CH C C10%
C50
C20%
sOUR
Cellulase activity
Chapter 5. Towards a competitive SSF
98
b) Sequential batch operation for cellulase production
The operation of bioreactors working under SB90 and SB50 conditions is presented in Figures 5.2 and
5.3 respectively. In both cases, an initial 2-day fermentation process was performed until the
maximum sOUR was achieved adding compost in 10% ratio as a mixed inoculum. Final average
sOUR and cellulases activity in these initial fermentations for both SB50 and SB90 strategies were
2.84±0.66 mgO2 g-1DM h-1and 7.6± 3.0 FPU g-1DM respectively. A similar temperature profile was
obtained, with a maximum of 62ºC in both cases. The characterization of the initial and final
fermented solids for both tested strategies is presented in Table 5.1.
Six sequential batches were performed (12 days of operation) for the SB90 strategy as presented in
Figure 5.2. Operation was consistent and maximum sOUR and final cellulase activity remained stable
among the different batches, with an average sOUR of 2.6 ±0.1 mgO2 g-1DM h-1and 9.0±0.8 FPU g-
1DM. No statistical differences were observed for cellulase activity among the six batches. The
average COC was 87 ±6 mgO2 g-1DM per batch (48h).
For SB50 strategy the fermentation was performed for a total process time of 18 days (Figure 5.3).
After the first substrate change the sOUR achieved at the end of the batch (24h) was 68% lower
although cellulase activity was statistically similar to that of the initial batch. After two batches,
maximum sOUR performed consistently to an average value of 3.4 ± 0.2 mgO2 g-1DM h-1 (alterations
in days 11 and 15 correspond to failures in the aeration system and the process quickly recovered
from this). The average COC was 58 ± 6 mgO2 g-1DM per batch (24h). Contrary to the sOUR
dynamics, cellulase production was not consistent. Cellulase activity of final extracts dropped
gradually to 2.1 ± 1.0 FPU g-1DM on day 10, to increase after day 13 to final values of 10 FPU g-1DM
on day 18, statistically similar to the yields obtained with the SB90 strategy.
Partial cellulose degradation of 3.1% and 3.2% was found for SB50 and SB90 in 24 and 48h
respectively, which is in line with the values found in literature. Hemicellulose degradation of nearly
2%was observed in both SB50 and SB90. Lignin degradation was 2.5% and 0.8% for SB50 and
SB90. These results are comparable to those reported by Umasaravanan et al. (2011) that observed
only a 6% cellulose reduction and 3.1% lignin reduction in 21 days using sugarcane bagasse and rice
straw. Other authors reported high cellulose degradation in SSF of lignocellulosic materials, for
instance, 18% in 3 days using olive wastes (Salgado et al., 2015) or 48% degradation in 23 days using
high-cellulose paper waste (Das et al., 1998).
Lack of standardized cellulase activity determination makes difficult to provide a proper comparison
with reported results. Even more, the most common substrate used for cellulase activity, filter paper,
generates great concerns on reproducibility and accuracy, especially in mediums with low β -
Chapter 5. Towards a competitive SSF
99
glucosidase production (Coward et al., 2003). However, cellulase activity obtained in this work is
located in the range of reported researches, between 1-25 FPU g-1DM for small bioreactors under
sterile and mesophilic conditions (Behera et al., 2016).
Figure 5.2.Process follow up of sequential batch operation SB90 of coffee husk for cellulase
production. In a) temperature and cellulase activity and b) sOUR profiles are presented. Bars that do
not share a letter are significantly different.
Chapter 5. Towards a competitive SSF
100
Figure 5.3. Process follow up of sequential batch operation SB50 of coffee husk for cellulase
production. In a) temperature and cellulase activity and b) sOUR profiles are presented. Bars that do
not share a letter are significantly different.
Chapter 5. Towards a competitive SSF
101
As it can be observed in Fig. 5.2 and 5.3, similar final cellulase yields were obtained for both
strategies. Also in both cases sOUR profile was displaced and the maximum was reached sooner in
comparison with the two first batches with compost. As previously discussed, the maximum cellulase
activity coincided with the maximum sOUR when using compost as inoculum.
According to sOUR profiles, a reduction in lag phase was accomplished and global biological activity
was higher with the specialized biomass obtained at the end of the sequential operation fermentation
(final batches) compared to the first batch using compost as inoculum. In general terms, a reduction of
the lag phase and an increase in biological activity were expected as a result of inoculation and
specialization of biomass (Jurado et al., 2015). This suggests growth stimulation of the microbial
communities due to the inoculation of well adapted microbiota. Moreover, it has been stated that
inoculation of mixed inoculums not only improved the process on its physical chemical parameters,
but also stimulated microbiota growth and microbial diversity during the process (Ishii et al., 2000,
Liu et al., 2011).
An enhancement of cellulases production was expected due to biomass specialization. Cellulase
production is regulated by an inhibition system, in which cellobiose and glucose inhibits β -
glucosidase (Kuhad et al., 2016). These readily metabolizable sugars released to the medium are
easily consumed by the microorganisms present in the fermentation, which is one of the advantages of
working with SSF. In this sense, it seems reasonable to assume that cellulase production was affected
by other circumstances, such as the release of inhibitory compounds generated as the results of lignin
degradation (Brijwani et al., 2011), non-productive adsorption to lignin hydrolysates (Akimkulova et
al., 2016) or even to depletion of different mineral sources (Salgado et al., 2015).
A single research has been found on sequential batch in SSF for lipids and cellulases production by
oleaginous fungi (Cheirsilp et al., 2015), but this study was performed in a different scale (<1g) and
under sterile conditions. However, the authors found that at different substrate exchange ratio (>50%),
cellulases production could be sustained in values ranging 1.8-2.3 FPU g-1DM, although no increment
on its activity was found, in accordance with the findings described in this work.
c) Cellulase production profile using specialized inoculum.
The final fermented solids with selected microbiota obtained for each strategy were stored frozen and
later used to inoculate one batch each (per triplicate). The complete cellulase, temperature and sOUR
profiles are presented in Figure 5.4. Maximum FPase of 9.8 ± 2.4 FPU g-1DM from SB90 and 4.8±1.1
FPU g-1DM from SB50 were reached at day 2 using inoculum. Although both systems reached
maximum sOUR values around 4 mgO2 g-1DM h-1,sOUR dynamics were different with both
Chapter 5. Towards a competitive SSF
102
inoculums. sOUR peaked sooner in SB50 (24h) while total oxygen consumed was similar in both
cases (137 ± 22 and 148 ± 38 mgO2 g-1DM for SB90 and SB50 respectively).
Figure 5.4. sOUR, temperature and filter paper activity profiles using as inoculum the fermented solid
obtained at the end of a) SB90 and b) SB50 operation.
In spite of that fact, maximum sOUR values were sustained during the fermentation, only starting to
decay after day two of operation. Also in both cases maximum cellulase activity was achieved during
the sOUR peak, reaching values of nearly 10 and 5 FPU g-1DM for SB90 and SB50 respectively. The
Time [d]0.0 0.5 1.0 1.5 2.0 2.5
sOU
R [m
g O
2 g-1
DM
h-1
]
0
2
4
6
8
10
Cel
lula
se a
ctiv
ity [F
PU g
-1D
M]
0
2
4
6
8
10
12
14
Tem
pera
ture
[ºC
]
0
10
20
30
40
50
60
70
Time [d]
0.0 0.5 1.0 1.5 2.0 2.5
sOU
R [m
g O
2 g-1
DM
h-1
]
0
2
4
6
8
10
Cel
lula
se a
ctiv
ity [F
PU g
-1D
M]
0
2
4
6
8
10
12
14
Tem
pera
ture
[ºC
]
0
10
20
30
40
50
60
70
sOUR FPase Temperature
a)
b)
Chapter 5. Towards a competitive SSF
103
fact that sOUR and enzymatic production profile presented slight differences comparing when using
compost as inoculum is a demonstration of the different dynamics of the new microbial population.
In this case, a reduction in lag phase was accomplished with the specialized biomass obtained at the
end of the fermentation compared to the first batch using compost as inoculum.
Both cellulases profiles obtained with the specialized inoculums presented values according to the
obtained during the fermentations, even when a smaller size inoculum was added; which was the case
of SB50.
As mentioned in section 5.1.3b, it is expected a reduction in the lag phase and an increase on the
duration of the biological activity when an inoculation is performed. This can be easily observed in
Figures 5.4a and b, where in both configuration the lag phase was reduced to less than 10h and
thermophilic conditions and maximum sOUR was sustained until the end of the batch.
d) Microbial communities characterization
Bacteria
A total of 22 bacterial families were detected in SB50 and SB90 with an abundance above 1% as
presented in Figure 5.5. Only those families present in C and CH that were also found in SB50 and
SB90 final products are depicted for C and CH in order to better illustrate the origin of the bacterial
community present in the final products. Other families that did not thrive along the fermentations
have been ignored for clarity purposes.
SB50 and SB90 presented a uniform distribution, with roughly 15 and 20% of less than 1% of relative
abundance, and a relative high biodiversity as indicated by Shannon-Wiener index, which reached
levels greater than 3 for this taxonomic level (3.43 and 3.07, respectively). Taking into account the
values found for both raw materials (3.75 for C and 1.73 for CH), bacterial communities in two final
products seem to be more influenced by compost, despite its lower percentage at the initial mixture.
Sorensen´s-Dice indices, qualitative and quantitative (Figure 5.6), strengthen this assessment, since
higher values and, consequently, greater similarities were obtained when comparisons were
established between either both final products and C.
According to these results and those obtained from preliminary assay at lab scale (Section 5.1.3a), CH
would be a proper and more efficient substrate to promote cellulase activity on account of its high
cellulose content, but the bacterial cellulolytic community in compost dominates the structure of the
final products. Compost is the result of a complex process characterized by the presence of
recalcitrant and non-readily degradable substrates.
Chapter 5. Towards a competitive SSF
104
Figure 5.5. Bacterial distributions at the family level according to the16s sequencing for Compost
(C), Coffee Husk (CH) and the final product of reactors SB50 and SB90. Only families with a relative
abundance >1% in either SB50 or SB90 are depicted. Families detected with a relative abundance <
1% in these two samples are grouped as "others".
From a biological point of view, this results in a very strong competitive selection of the
microorganisms able to carry out all the necessary transformations. Thus, a concentration effect takes
place throughout the process (Lopez-Gonzalez et al., 2014) that promotes the presence of a more
specialized microbiota in the final product, adapted to those specific nutritional conditions that are
typical in composting, among them, cellulolytic microorganisms.
The lower similarity between the bacterial population associated to CH and both final products is
clearly evidenced by the fact that the most abundant families in CH, Pseudomonadaceae (33.7%),
Leuconostocaceae (22.6%), Enterobacteriaceae (17.7%) and Enterococcaceae (16.3%) (Table S1),
were sparingly represented in the latter. Thus, only the latest was found with a relative presence over
1% in SB50. On the other hand, in C nearly 80% of all families were a part of Firmicutes,
Proteobacteria and Actinobacteria phyla, which have been found to be predominant in compost from
lignocellulosic materials (Zhang et al., 2016). Clostridiaceae was the most predominant family in C
present in 7.7%, which was present in both SB50 and SB90. It is important to point out that
Pseudomonaceae and Rhodospirillaceae families were more abundant in the CH and C respectively,
Chapter 5. Towards a competitive SSF
105
however their species were not able to thrive at the fermentations conditions of SB50 and SB90.
Amounts of these families on final fermentation products were below 1%.
Figure 5.6. Sorensen´s-Dice indices for prokaryotic (black) and eukaryotic (grey) microbiota
associated to raw materials and final products (Compost (C), Coffee Husk (CH) and the final product
of reactors SB50 and SB90). Every quadrant shows both qualitative ( , ) and quantitative ( , )
indices for all taxonomic levels (from left to right: phylum, family, genus and specie).
SB50 and SB90 resulted in similar bacterial distribution, in which compost influence as inoculum is
clearly reflected. Sphingobacteriaceae, Paenibacillaceae and Xanthomonadaceae families together
were present in a discrete 2.2% in C, however, they represented a 47.2 and 33.1% in SB50 and SB90
respectively. From these families, several species have been identified during the early stage of
lignocellulosic material composting process (Lopez-Gonzalez et al., 2015a).
From the dominant families, Pseudoxanthonomas taiwanensis and Sphingobacterium composti
appeared as most abundant identified species in SB50 and SB90. Relative abundance of these species
were 12.3 and 6.1% for SB50 and 6.1 and 2.6% for SB90 respectively. P. taiwanensis has been
C CH SB90 SB50
C
CH
SB90
SB50
0
0
0
01
1
1
1
Chapter 5. Towards a competitive SSF
106
widely related to cellulose degradation systems, for its β -glucosidase production and the potential
enhancement of the growth of other cellulolytic bacterias (Eichorst et al., 2013). On the other hand, S.
composti has been related to lignin degradation and the production of acids in aerobic conditions
(Karadag et al., 2013). The presence of these two species confirmed the results obtained in both
sequential batch operations, which presented significant lignin and cellulose degradation.
Furthermore, from Paenibacillaceae family Paenibacillus and Thermobacillus genera were the most
abundant. Several species from both genera have been widely reported as strong cellulose degraders
such as P.chitinolyticus (Mihajlovski et al., 2015), which was found on SB50 and SB90 in near 2% of
relative abundance. Also T.composti has been found in 4 and 2.5% in SB50 and SB90 respectively.
This specie is associated with cellobiase and xylanolytic activity during thermophilic stage in
composting processes (de Gannes et al., 2013). Lately it has been characterized as halotolerant, which
is of great interest for its bioethanol potential using ionic liquids (Watanabe et al., 2007).
Fungi and yeasts
A total of 12 families were detected for C, CH, SB50 and SB90 as presented in Figure 5.7. First of all,
it is possible to notice that in all samples there is a great amount of non-identified DNA, described
only as Eukariota, representing 25.7, 5.5, 44.8 and 5.8% of mycobiota from C, CH, SB50 and SB90
respectively. This relatively high rate of unidentified fungi may be consequence of the still scarce and
sometimes confusing information available on data base concerning molecular characterization of
fungi, especially in comparison with bacteria, which now limits the potential for elucidating the
structure of the mycobiota associated to different environments (Langarica-Fuentes et al., 2014).
Remarkably, most of the fungi and yeasts in SB50 and SB90 come from CH where four families
account for 94.4% of all mycobiota. This similarity was particularly notable for SB50, with
remarkable differences as expressed by the values of the Sorensen´s-Dice index associated to both
comparisons, SB50 v. C and SB50 v. CH (Figure 6). Fungal biodiversity increased in the final
materials compared to CH as the Shannon indices obtained were 1.63, 0.98, 1.39 and 1.33 for C, CH,
SB50 and SB90. Predominant families were Phaffomycetaceae, Dipodascaceae, and two unidentified
families of the class of Tramellomycetes. Distribution in SB50 and SB90 was similar in content but
not in proportion, resulting in higher qualitative homology in contrast to quantitative.
Phaffomycetaceae family was present in CH in 7.8% and in negligible amounts in C. This family
thrived on both reactors, achieving a relative abundance of 21.9 and 43.2% in SB50 and SB90
respectively. In this family two yeasts species were identified.
Chapter 5. Towards a competitive SSF
107
Figure 5.7. Identified fungal and yeasts families in Compost (C), Coffee Husk (CH) and the final
product of reactors SB50 and SB90.
The most abundant specie was identified as Cyberlindnera jardinii,(17.8 and 34.0% in SB50 and
SB90 respectively), which is able to metabolize pentoses and tolerates lignin by-products, which
makes it a suitable specie for the treatment of lignocellulosic wastes (Nordberg et al., 2014, Lopez-
Gonzalez et al., 2015b). The second specie was Barnettozyma californica (4.1 and 9.1% in SB50 and
SB90 respectively), a xylose-fermenting and xylanase producer yeast (Morais et al., 2013). In this
sense, in SB90, most of fungal/yeast population is focused on hemicellulases metabolism which, in a
way, could increase cellulose availability by relaxing the rigid lignocellulosic matrix, which would
result in an easier accessibility to this macropolymer, and an increase in the cellulases production.
Another difference between both final fermented solids is the high abundance of Dipodascaceae
family only in SB50, with a 21.9%. in comparison with 0.15% of SB90. Dipodascus australiensis and
Galactomyces sp. are the most representative yeasts of this family and has been proven to be xylanase
producer and the latter, cellulose degrader with mild halotolerant characteristics (He et al., 2016).
Finally in Tremellomycetes class, Tremellales order, 3 different genera were detected as
Trichosporonales sp. LM659, uncultured Tremellales and uncultured Trichosporon which all together
account as 85.5, 10.3 and 46.5% of CH, SB50 and SB90 respectively. No species were identified in
this order, however some reports on different species of this order has been proven to be
unconventional due to absence of diauxic effect, i.ef ermenting hexoses and pentoses while also
Chapter 5. Towards a competitive SSF
108
effectively utilizes xylose and N-acetylglucosamine, which are building blocks of lignocellulosic
materials (Kourist et al., 2015).
Overall performance of the process and the microbial communities involved
In general, solids obtained at the end of SB50 and SB90 fermentations, provided a strongly
specialized mixed inoculum for lignocellulosic materials degradation. Cellulose degrading activity
was mainly provided by bacterial populations; however it is important to remark that cellulolytic
bacteria do not always express the full cellulose degradation system (Behera et al., 2016). On the
other hand, hemicellulose degradation potential was provided by fungal/yeasts species. In spite of the
abundance of hemicellulose degraders, hemicellulose degradation was negligible (Table 1). It is
possible that this biomass was not active during the process, or the activity was low in comparison
with cellulose or even lignin degraders, or that the hemicellulolytic microorganisms act partially on
the heteropolymer, promoting relaxation but not degradation in an extensive rate.
Considering these results it is likely that synergistic or complementary actions take place in the total
microbiota of the reactors. Degradation of cellulose does not occur in isolation. As a constituent of the
lignocellulosic matrix, the action of cellulases is not the only event needed for the transformation of
this glucose polymer (Kanokratana et al., 2015). Thus, the biodegradation process is the results of
different cooperative actions that first promote the relaxation of the enmeshed structure that
characterize lignocellulose. This initial event allows the access of enzymes to the site of action and,
consequently, is one of the key steps in order to achieve the depolymerisation, not just of cellulose but
of all macropolymers in lignocellulose too (Duarte et al., 2012). From a microbial point of view, the
complexity of the process acts as a pressure factor that promotes the selection for a microbiota
metabolically adapted to the nutritional demands associated to lignocellulosic environments. This is
particularly true for artificial habitats, as in the case here described, in which microbiome shifts from
a metabolically diverse community to a specialized microbiota (López-González et al., 2014). In this
sense, it would be highly improbable that lignocellulosic and lignocellulosic-inhibiting populations
coexist, after the occurrence of such selective process. Nevertheless, it cannot be excluded some other
inhibitory effects associated to the products resulting from the action of cellulases and other
lignocellulosic enzymes, as glucose, cellobiose or phenolic compounds, although microorganisms that
consume these inhibitory compounds use to be members of the lignocellulosic community
(Wongwilaiwalin et al., 2013).
In light of the experimental data presented in this work, it is possible to remark that the use of
fermented solids as inoculum in a sequential batch operation is very successful. The proposed
operation offers a way to reduce costs in inoculum requirements and potentially to the reduction of
Chapter 5. Towards a competitive SSF
109
solid wastes generation as suggested by Farinas (2015). The studies on microbial diversity on non-
sterile SSF processes are scarce. In this process a specialized biomass provides the generation of an
enzymatic pool, containing different enzymatic activities related to lignocellulose degradation. The
positive effects of using a multi-enzymatic preparation have been reported by Melikoglu et al., (2013).
These authors were able to improve bioethanol production by producing several enzymes by SSF.
Furthermore, it has been proven that the use of these enzymatic pools without any additional
processing steps provides better results than commercial preparations, reducing operational costs
(Lever et al., 2005). In summary, the proposed process presents the following economic advantages:
the use of organic wastes instead of pure substrates; saving the investment and the operating costs
related to sterilization; saving the costs of producing inoculum for each batch, since fermented solids
from one batch are re-used to inoculate the following batch; the potential use of the multienzymatic
extracts without additional purification steps. Further economic assessment should confirm the cost
effectiveness of the process.
5.1.4 Conclusion
Operation of SSF of coffee husk in sequential batches has been proven as a suitable strategy for
cellulases production using a mixed inoculum. SB50 and SB90 strategies provided a sustained
fermentation for 18 and 12 days respectively and cellulase production stabilised at around 10 FPU g-
1DM. 48h batches in the strategy SB90 provided a more consistent operation while SB50 required
more time to reach a pseudo-steady state. Both strategies obtained cellulase activity in the reported
range of production. The sequential process allowed the enrichment of cellulose and hemicellulose
degraders, eliminating the requirements of fresh inoculum for each batch. Bacterial communities
obtained at the end of both processes came from compost with great cellulose degradation potential.
Fungal and yeasts communities came mostly from coffee husk, with high hemicellulose degradation
potential. The development of these operational strategies and further biological characterization of
the end product could eventually benefit the process economics by providing a standard and
specialized inoculum for a continuous solid state fermentation for cellulases production.
References
Abraham, J., Gea, T., Sánchez, A. 2013. Potential of the solid-state fermentation of soy fibre residues by native microbial populations for bench-scale alkaline protease production. Biochemical Engineering Journal. 74,15-9. Akimkulova, A., Zhou, Y., Zhao, X., Liu, D. 2016. Improving the enzymatic hydrolysis of dilute acid pretreated wheat straw by metal ion blocking of non-productive cellulase adsorption on lignin. Bioresource Technology. 208, 110-6.
Chapter 5. Towards a competitive SSF
110
Arora, R., Behera, S., Kumar, S. 2015. Bioprospecting thermophilic/thermotolerant microbes for production of lignocellulosic ethanol: A future perspective. Renewable Sustainable Energy Reviews. 51,699-717. Astolfi, V., Joris, J., Verlindo, R., Oliveira, J., Maugeri, F., Mazutti, M., De Olivera, D., Treichel, H. 2011. Operation of a fixed-bed bioreactor in batch and fed-batch modes for production of inulinase by solid-state fermentation. Biochemical Engineering Journal. 58, 39-49. Behera, S., Ray, R.2016. Solid state fermentation for production of microbial cellulases: Recent advances and improvement strategies. International Journal of Biological Macromolecules.86, 656- 69. Brijwani, K., Vadlani, P. 2011. Cellulolytic enzymes production via solid-state fermentation: Effect of pretreatment methods on physicochemical characteristics of substrate. Enzyme Research. 2011,860134. Cerda, A., El-Bakry, M., Gea, T., Sánchez, A. 2016. Long term enhanced solid-state fermentation: Inoculation strategies for amylase production from soy and bread wastes by Thermomyces sp. in a sequential batch operation. Journal of Environmental and Chemical Engineering. 4,2394- 2401. Cheirsilp, B., Kitcha, S. 2015. Solid state fermentation by cellulolytic oleaginous fungi for direct conversion of lignocellulosic biomass into lipids: Fed-batch and repeated-batch fermentations. Industrial Crops and Products. 66,73-80. Coward-Kelly, G., Aiello-Mazzari, C., Kim, S., Granda, C., Holtzapple, M. 2003. Suggested improvements to the standard filter paper assay used to measure cellulase activity. Biotechnology and Bioengineering. 82:745-9. Das, K., Tollner, E., Tornabene, T. 1998. Composting pulp and paper industry solid wastes: process design and product evaluation. Proceedings of the composting in the southeast-conference and expo Athens, GA. p. 9-11. de Gannes, V., Eudoxie, G., Hickey, W. 2013. Prokaryotic successions and diversity in composts as revealed by 454-pyrosequencing. Bioresource Technology. 133, 573-80. Dhillon, G., Kaur, S., Brar, S., Verma, M. 2012. Potential of apple pomace as a solid substrate for fungal cellulase and hemicellulasebioproduction through solid-state fermentation. Industrial Crops and Products. 38, 6-13. Duarte, G.C, Moreira, L.R.S., Jaramillo, P.M.D., Filho, E.X.F. 2012. Biomass-derived inhibitors of holocellulases. Bioenergy Research. 5, 768-777. Eichorst, S., Varanasi, P., Stavila, V., Zemla, M., Auer, M., Singh, S. 2013. Community dynamics of cellulose-adapted thermophilic bacterial consortia. Environmental Microbiology. 15, 2573-87. El-Bakry, M., Abraham, J., Cerda, A., Barrena, R., Ponsá, S., Gea, T., Sánchez, A. 2015. From Wastes to High Value Added Products: Novel Aspects of SSF in the Production of Enzymes. Critical Reviews in Environmental Science and Technology. 45,1999-2042. El-Bakry, M., Gea, T., Sánchez, A. 2016. Inoculation effect of thermophilic microorganisms on protease production through solid-state fermentation under non-sterile conditions at lab and bench scale (SSF). Bioprocess and Biosystems Engineering. 39,585-92. Farinas C. 2015. Developments in solid-state fermentation for the production of biomass degrading enzymes for the bioenergy sector. Renewable and Sustainable Energy Reviews. 52, 179–188. Garcia-Ochoa, F., Gomez, E., Santos, V., Merchuk, J. 2010. Oxygen uptake rate in microbial processes: An overview. Biochemical. Engineering. Journal. 49, 289-307. Ghose T. 1987. Measurement of cellulase activities. Pure Applied Chemistry. 59,257-268. He, Y., Liu, F., Gong, L., Di, J., Ding, Y., Ma, C. 2016. Enzymatic in situ saccharification of chestnut shell with high ionic liquid-tolerant cellulases from Galactomyces sp. CCZU11-1 in a biocompatible ionic liquid-cellulase media. Bioresource Technology. 201,133-139. Heather, L., Sabrina, S. 2009. The contribution of enzymes and process chemicals to the life cycle of ethanol. Environmental Research Letters. 4,14001. Ishii, K., Fukui, M., Takii, S. 2000. Microbial succession during a composting process as evaluated by denaturing gradient gel electrophoresis analysis. Journal of Applied Microbiology. 89, 768- 77.
Chapter 5. Towards a competitive SSF
111
Jurado, M., Suárez-Estrella, F., Vargas-García, M., López, M., López-González, J., Moreno, J. 2014. Evolution of enzymatic activities and carbon fractions throughout composting of plant waste. Journal of Environmental Management. 133, 355-364. Jurado, M., Suárez-Estrella, F., López, M., Vargas-García, M., López-González, J., Moreno, J. 2015. Enhanced turnover of organic matter fractions by microbial stimulation during lignocellulosic waste composting. Bioresource Technology. 186, 15-24. Kanokratana, P., Eurwilaichitr, L., Pootanakit, K., Champreda, V 2015. Identification of glycosyl hydrolases from a metagenomic library of microflora in sugarcane bagasse collection site and their cooperative action on cellulose degradation. Journal of Bioscience and Bioengineering. 4, 384-391. Karadag, D., Özkaya, B., Ölmez, E., Nissilä, M., Çakmakç, M., Yıldız, Ş., Puhakka, J. 2013. Profiling of bacterial community in a full-scale aerobic composting plant. International Biodeterioration and Biodegradation. 77, 85-90. Kourist, R., Bracharz, F., Lorenzen, J., Kracht, O., Chovatia, M., Daum, C. 2015. Genomics and transcriptomics analyses of the oil-accumulating basidiomycete yeast Trichosporon oleaginosus: Insights into substrate utilization and alternative evolutionary trajectories of fungal mating systems. mBio. 6, 00918-15. Kuhad, R., Deswal, D., Sharma, S., Bhattacharya, A., Jain, K., Kaur, A. 2016. Revisiting cellulase production and redefining current strategies based on major challenges. Renewable and Sustainable Energy Reviews.55, 249-72. Langarica-Fuentes, A., Zafar, U., Heyworth, A., Brown, T., Fox, G., Robson, G. 2014. Fungal succession in an in-vessel composting system characterized of using 454 pyrosequencing. FEMS Microbiology Ecology. 88, 296-308. Lever M. 2005. Modelling the energy performance of a farm-scale cellulose to ethanol process with on-site cellulase production and anaerobic digestion. Renewable Energy. 74,893-902. Lever, M., Ho, G., Cord-Ruwisch, R. 2010. Ethanol from lignocellulose using crude unprocessed cellulase from solid-state fermentation. Bioresource Technology. 101, 7083–7087 Liu, J., Xu, X., Li, H., Xu, Y. 2011. Effect of microbiological inocula on chemical and physical properties and microbial community of cow manure compost. Biomass and Bioenergy. 35, 3433-9. López-González, J., Vargas-García, M., López, M., Suárez-Estrella, F., Jurado, M., Moreno, J. 2014. Enzymatic characterization of microbial isolates from lignocellulose waste composting: Chronological evolution. Journal of Environmental Management. 145, 137-46. López-González, J., Suárez-Estrella, F., Vargas-García, M., Lopez, M., Jurado, M., Moreno, J. 2015a. Dynamics of bacterial microbiota during lignocellulosic waste composting: Studies upon its structure, functionality and biodiversity. Bioresource Technology. 175, 406-416. López-González, J., Vargas-García, M., López, M., Jurado, M., Moreno, J. 2015b. Biodiversity and succession of mycobiota associated to agricultural lignocellulosic waste-based composting. Bioresource Technology. 187, 305-313 McMillan, J., Jennings, E., Mohagheghi, A., Zuccarello, M. 2011. Comparative performance of precommercial cellulases hydrolyzing pretreated corn stover. Biotechnology for Biofuels. 4,1- 17. Melikoglu, M., Sze, C., Webb, C. 2013. Kinetic studies on the multi-enzyme solution produced via solid state fermentation of waste bread by Aspergillus awamori. Biochemical Engineering Journal. 80, 76–82. Mihajlovski, K., Carević, M., Dević, M., Šiler-Marinković, S., Rajilić-Stojanović, M., Dimitrijević- Branković, S. 2015. Lignocellulosic waste material as substrate for Avicelase production by a new strain of Paenibacillus chitinolyticus CKS1. International Biodeterioration and Biodegradation. 104,426-34. Miller G. 1959. Use of dinitrosalicylic acid reagent for determination of reducing sugar. Analytical Chemistry. 31,426-8. Mitchell, D., Berovič, M., Krieger, N. 2006. Solid-State Fermentation Bioreactor Fundamentals: Introduction and Overview. In: Mitchell, D., Berovič, M., Krieger, N., editors. Solid-State
Chapter 5. Towards a competitive SSF
112
Fermentation Bioreactors: Fundamentals of Design and Operation. Berlin, Heidelberg: Springer Berlin Heidelberg; p. 1-12. Morais, C., Cadete, R, Uetanabaro, A., Rosa, L., Lachance, M., Rosa, C. 2013. D-xylose-fermenting and xylanase-producing yeast species from rotting wood of two Atlantic Rainforest habitats in Brazil. Fungal Genet Biol. 60,19-28. Murthy, P.S., Naidu, M.M. 2012. Sustainable management of coffee industry by-products and value addition—A review. Resources, Conservation and Recycling. 66, 45–58. Mussatto, S. I., Machado, E. M. S., Martins, S., Teixeira, J. A. 2011. Production, composition, and application of coffee and its industrial residues. Food Bioprocess Technology. 4, 661–672. Narita, Y., Inouye, K. 2014. Review on utilization and composition of coffee silverskin. Food Research International. 61, 16-22. Nordberg, H., Cantor, M., Dusheyko, S., Hua, S., Poliakov, A., Shabalov, I. 2014. The genome portal of the Department of Energy Joint Genome Institute: 2014 updates. Nucleic Acids Research. 42:D26-D31. Pensupa, N., Jin, M., Kokolski, M., Archer, D. Du, C. 2013. A solid state fungal fermentation-based strategy for the hydrolysis of wheat straw. Bioresource Technology.149, 261-267 Pessoa, D., Finkler, A., Machado, A., Luz, L., Mitchell D. 2016. Fluid dynamics simulation of a pilot- scale solid-state fermentation bioreactor. Chemical Engineering Transactions. 49: 49-54. Olofsson, K., Bertilsson, M., Lidén, G.2008. A short review on SSF – an interesting process option for ethanol production from lignocellulosic feedstocks. Biotechnology for Biofuels. 1, 1-14. Ponsá, S., Gea, T., Sánchez, A. 2010. Different Indices to Express Biodegradability in Organic Solid Wastes. Journal of Environmental Quality. 39, 706-12. Ruggieri, L., Gea, T., Artola, A., Sánchez, A. 2009. Air filled porosity measurements by air pycnometry in the composting process: a review and a correlation analysis. Bioresource Technology. 100, 2655–2666. Salgado, J., Abrunhosa, L., Venâncio, A., Domínguez, J., Belo, I. 2015. Enhancing the bioconversion of winery and olive mill waste mixtures into lignocellulolytic enzymes and animal feed by Aspergillus uvarumusing a packed-bed bioreactor. Journal of Agricultural and Food Chemistry. 63, 9306-14. Shemekite, F., Gómez-Brandón, M., Franke-Whittle, I.H., Praehauser, B., Insam, H., Assefa, F. 2014. Coffee husk composting: An investigation of the process using molecular and non-molecular tools. Waste Management. 34, 642-652. Tolan, J., Foody, B. 1999. Cellulase from submerged fermentation. In: Tsao GT, Brainard AP, Bungay HR, Cao NJ, Cen P, Chen Z, editors. Recent Progress in Bioconversion of Lignocellulosics. Berlin, Heidelberg: Springer Berlin Heidelberg;p. 41-67. Umasaravanan, D., Jayapriya, J., Rajendran, R. 2011. Comparison of lignocellulose biodegradation in solid state fermentation of sugarcane bagasse and rice straw by Aspergillus tamarii. Ceylon Journal of Science (Biological Sciences). 40 (1), 65-68. US Composting Council. 2001.Test methods for the examination of composting and compost. Houston: Edaphos International. Van Soest, P., Robertson, J., Lewis, B. 1991. Methods for dietary fiber, neutral detergent fiber, and non starch polysaccharides in relation to animal nutrition. Journal of Dairy Science. 74(10), 3583-3597. Watanabe, K., Nagao, N., Yamamoto, S., Toda, T., Kurosawa, N. 2007. Thermobacillus composti sp. nov., a moderately thermophilic bacterium isolated from a composting reactor. International Journal of Systematic and Evolutionary Microbiology. 57, 1473-7. Wongwilaiwalin, S., Laothanachareon, T., Mhuantong, W., Tangphatsornruang, S., Eurwilaichitr, L., Igarashi, Y. 2013. Comparative metagenomic analysis of microcosm structures and lignocellulolytic enzyme systems of symbiotic biomass-degrading consortia. Applied Microbiology and Biotechnology. 97, 8941–8954.
Chapter 5. Towards a competitive SSF
113
Zhang, L., Zhang, H., Wang, Z., Chen, G., Wang, L. 2016. Dynamic changes of the dominant functioning microbial community in the compost of a 90-m3 aerobic solid state fermentor revealed by integrated meta-omics. Bioresource Technology. 203, 1-10.
Chapter 5. Towards a competitive SSF
114
Chapter 6. Reproducibility assessment
115
CHAPTER 6. REPRODUCIBILITY ASSESSMENT
OF CELLULASE PRODUCTION THROUGH
SOLID-STATE FERMENTATION
Part of this chapter is submitted for publication to Biotechnology for Biofuels. "Cellulase and
xylanase production at pilot scale by solid-state fermentation from coffee husk using specialized
biomass: Process operation and microbial communities consistency."
Cerda, A., Mejias, L., Gea, T., Sánchez, A.
Chapter 6. Reproducibility assessment
116
Chapter 6. Reproducibility assessment
117
Summary
This chapter summarizes and discusses the reproducibility and of three SSF, performed in 50-L
reactors using coffee husk as the substrate and a previously propagated specialized inoculum.
This study has two main sections, one is the study of the three different fermentations carried out in
4.5L reactors. These fermentations performed as propagation for the specialized inoculum that was
used for the pilot fermentations. The other section is the complete study of the pilot fermentations, in
terms of biological activity parameters, enzymatic activity, gaseous emissions and microbial
communities monitoring. As mentioned in other chapters, solid wastes mixtures are heterogeneous in
comparison to liquid mediums used in conventional fermentations; in addition, these processed were
inoculated with a fermented solid coming from another dynamic and complex process. Considering
these aspects and looking for an standard and consistent SSF, the objective of this compilation is to
know how this heterogeneity influences the productive processes and if these processes are
environmentally friendly. There is a great lack of information regarding SSF using non sterile
substrate, for this reason, the conclusions of this work will provide useful information for further SSF
standardization. This work was carried out in collaboration with Laura Mejías.
6.1 Process operation and microbial communities consistency in a solid state fermentation for
cellulase and xylanase production.
6.1.1 Materials
Coffee husk (CH) was kindly provided by Marcilla S.A (Montcada, Barcelona, Spain) and stored
frozen at -20ºC until use. CH presented a general characterization of: moisture content of 60.8±0.4%
(wet basis), pH 6.5±0.1 and C/N ratio of 13.3±0.5. Fiber content in CH was 25.7±0.2, 14.6±0.1 and
17.6±0.5% (dry basis) of cellulose, hemicellulose and lignin, respectively.
The inoculum used in this work consisted of a specialized consortium of microorganisms able to
produce cellulases using CH as sole substrate. This was previously obtained operating SSF by
sequential batch (SB) using compost as starting inoculum. Main identified species were the bacteria
Pseudoxanthonomas taiwanensis and Sphingobacterium composti and the yeasts Cyberlindnera
jardinii and Barnettozyma californica, all of them previously reported as lignocellulose degraders.
The procedure to obtain this inoculum is described in Chapter 5. Briefly, compost was added in a 10%
(w/w) ratio to a mixture of CH and wood chips as bulking agent in a 1:1 (v/v) ratio to provide the
proper porosity to the mixture (Ruggieri et al., 2009). During the entire process there was no evidence
Chapter 6. Reproducibility assessment
118
suggesting wood chips degradation due to the short time of each cycle. Sequential batch operation
was performed in cycles until stable cellulase production of 10 FPU g-1 DM (filter paper units per
grams of dry matter). SB operation started with a 48h fermentation stage followed by a substrate
exchange of 90% of volume every 48h. The process was operated for several cycles until stable
operation was achieved and the final fermented solids were stored frozen for further use as specialized
inoculum. A full characterization of the specialized consortium was performed by DNA sequencing
and the complete characterization is presented in Chapter 5 section 5.1.4d. The specialized inoculum
presented a general characterization of: moisture content of 55.0±3.1% (wet basis), pH 9.1±0.1 and
C/N ratio of 13.1±0.6. Fiber content in the specialized inoculum was 23.5±0.5, 13.3±0.5 and
18.3±0.6% (dry basis) of cellulose, hemicellulose and lignin, respectively.
6.1.2 Experimental procedure
SSF was performed in 50 L pilot bioreactors in triplicate batches (R1, R2 and R3), the complete
system is described in chapter 3 section 3.2.3. Each batch was inoculated with fermented solids from a
4.5 L propagation reactor (P1, P2 and P3, respectively) inoculated with the thawed specialized
consortium. The experimental set up used for the propagation reactors is describes in chapter 3 section
3.2.2. Both propagation and pilot operations were carried out under near adiabatic conditions, using
thermally isolated reactors in all cases. Full description is presented below
a) Inoculum preparation in propagation reactors
A propagation reactor was required to provide the proper amount of biomass to inoculate the pilot
reactors. Inoculum was obtained by using an initial mixture containing CH and 10% of the mixed
specialized inoculum with wood chips, which was used in a ratio of 1:1 (v/v), giving a total weight of
1.2 kg per batch. The entire process was carried out for 48h in 4.5 L air-tight reactors, working under
near adiabatic conditions and oxygen controlled aeration (120-240 mL min-1, oxygen setpoint 11.5%
in air) as described in chapter 3 section 3.4.1.
b) Pilot reactor operation
Three batches were performed in 50 L closed reactors. A schematic diagram of the pilot reactor and a
detailed description can be found in chapter 3 section 3.2.3 (Puyuelo et al., 2010). Temperature,
exhaust gas oxygen concentration and inlet airflow were monitored during the trials. The experiments
were performed with forced aeration and airflow was manually adjusted to ensure that the oxygen
content in the reactor remained above 10%, in order to provide full aerobic conditions (Puyuelo et al.,
2010).
Chapter 6. Reproducibility assessment
119
The mixtures were prepared by mixing CH and the specialized inoculum obtained from the
propagation reactor in a 90:10 (w/w) ratio respectively. Wood chips were added as bulking agent in a
volume ratio of 1:1 (v/v). The final weight of the mixture was 15.2 kg for each reactor. The first batch
(R1) was performed to obtain a full profile of enzymatic activity production, collecting gaseous and
solid samples at 0, 8, 16, 24, 35, 48, 58, 72 and 134 h, according to the methods described in section
3.3.5. In order to assess the reproducibility of the process, two additional batches (R2 and R3) were
performed until the moment of maximum cellulase activity.
c) Sampling
Gaseous samples were collected in 1-L Tedlar® bags for ammonia (NH3), volatile organic compounds
(VOC), nitrous oxide (N2O) and methane (CH4) content determination before opening of the reactor,
according to the methods described in chapter 3 section 3.3.5. Once the reactor was opened and after
homogenization, solid samples were collected for the determination of cellulase and xylanase
activities. Filter paper activity (FPase), carboxymethylcellulase (CMCase) and β-glucosidase (BGase)
activities were measured for cellulase production. Xylanase (Xyl) production was followed for
hemicellulase production.
In addition to the enzymatic measurements, the solid samples were analyzed in order to determine the
neutral detergent fiber, acid detergent fiber and lignin content. These analysis were performed
according to the methods presented in section 3.4. Degradation percentage of cellulose, hemicellulose
and lignin were calculated according to a mass balance and considering the weight evolution
throughout the process.
6.1.3. Results
a) Specialized inoculum preparation
Figure 6.1 presents the sOUR and the temperature profiles of the three propagation reactors (P1, P2
and P3) for inoculum preparation. It is possible to observe that a similar trend is presented in all three
propagation reactors. However, the heterogeneity of the materials led to a series of small differences
during the processes despite of using the same specialized consortium.
P2 and P3 started at temperatures of 21 and 25ºC respectively, which was notably higher than initial
P1 temperature (12ºC, Figure 6.1b). This difference was attributed to the preparation process of the
substrate, which includes a defrosting stage. Major differences were found during transition to
thermophilic stage after 24h of operation. After reaching 45ºC the automatic control started to act
Chapter 6. Reproducibility assessment
120
according to the oxygen requirements of the fermentation, which can be clearly observed in the
oscillation of sOUR profiles of P2 and P3. Control actuation reflects high biological activity in these
two reactors in detriment to P1, where sOUR only started to oscillate at the end of the fermentation. In
this sense, at the end of the fermentation sOUR obtained for P1, P2 and P3 were 2.6, 3.0 and 3.1 mg
O2 g-1 DM h-1 for P1, P2 and P3, respectively.
Figure 6.1. Operation profile of a) sOUR and b) temperature of the propagation reactors P1, P2 and
P3 (4.5L) of a specialized inoculum using coffee husk as substrate.
Additionally, COC resulted in very similar with values of 86.1 and 94.8 mg O2 g-1 DM for P2 and P3
respectively. In contrast, P1 presented a COC of 65.3 mg O2 g-1 DM, which represents around 70% of
the COC of P2 and P3. It is highly possible that initial temperature affected the performance of P1, in
detriment of biological activity (Sundh & Rönn, 2002). It is for this reason that, in order to improve
reproducibility of the process, in terms of biological and enzymatic activity, the initial temperature of
the process must be set on proper values to reduce the lag phase of microbial communities that will
grow on the reactors.
Cellulase activity produced in the three reactors was 8.1±0.2, 4.6±0.1 and 8.4±0.2 FPU g-1 DM for P1,
P2 and P3, respectively. This specialized inoculum was obtained in a SSF process performed by
sequential batch operation reaching consistent cellulases production within 8-9 FPU g-1 DM (see
chapter 5, section 5.1.3b). In this sense, the enzymatic activity observed in P2 is an unexpected value.
Further research should focus on finding correlations between specific operational parameters and
enzymatic activity to predict it.
Time [h]
0 10 20 30 40
sOU
R [m
gO2
g-1 D
M h
-1]
0
1
2
3
4P1P2P3
Time [h]
0 10 20 30 40
Tem
pera
ture
[ºC
]
0
10
20
30
40
50
60
70P1P2P3
a) b)
Chapter 6. Reproducibility assessment
121
b) Pilot reactor operation
Figure 6.2 presents the fermentation profile for a 134 h operation using a previously propagated
inoculum. The fermentation presented a quick start-up with no lag phase as observed in the sOUR and
temperature thermophilic profiles (Figure 6.2a). This was probably due to the initial conditions of the
inoculum, which was in its highest biological and enzymatic activity.
Figure 6.2. Operation profile of a) sOUR and Temperature and b) FPase, CMCase, BGase and Xyl
production profile in a pilot reactor (50L) using a propagated specialized inoculum and coffee husk as
Time [h]0 20 40 60 80 100 120 140
sOU
R [m
gO2 g
-1 D
M h
-1]
0
2
4
6
8
10
12
Tem
pera
ture
[ºC
]
0
20
40
60
80sOURTemperature
Time [h]
0 20 40 60 80 100 120 140
Xyl
anas
e ac
tivity
[U g
-1D
M]
0
10
20
30
40
50
Enzy
mat
ic a
ctiv
ity [U
g-1
DM
]
0
1
2
3
4
5
6
7Xyl FPaseCMCase BGase
a)
b)
Chapter 6. Reproducibility assessment
122
substrate, during 134 h. All enzymatic activities were expressed as U g-1DM with exception of FPase,
which is expressed as FPU g-1DM.
Biological activity was found at its maximum during the first 30h of fermentation in a full
thermophilic stage. Maximum sOUR and temperature were 4.1 mg O2 g-1 DM h-1 and 71ºC,
respectively, at 24h. Oxygen requirements reached a total COC of 243.2 mg O2 g-1 DM. COC found at
24h of fermentation, at its highest sOUR, was 80.2 mg O2 g-1 DM, which roughly represents 33% of
the total oxygen consumption. It is important to highlight the importance of choosing the exact
moment to have the maximum enzymatic activity, which in occasions, can match with the maximum
biological activity, reflected as sOUR (Abraham et al., 2013) or not (Cerda et al., 2016).
In addition, pH reached an alkaline value of 9.22±0.01 and a C/N ratio of 12.99±0.15, which reflected
organic matter degradation. These conditions remained until the end of the fermentation, achieving
final values of 9.34±0.01 for pH and a C/N ratio of 11.43±1.03.
Cellulolytic and hemicellulolytic activities also showed their maximum during the most active stage
of the fermentation (Figure 6.2b). FPase, BGase and Xyl reached their maximum production in the
first 24h of operation with values of 3.6±0.2 FPU g-1 DM, 0.12±0.01 U g-1 DM and 32±5 U g-1 DM,
respectively. FPase and BGase presented a local maximum, decreasing after this moment to negligible
values. This can be attributed to several factors such as nutrient depletion or the inhibition of the
enzymatic system due to the formation of by-products (Brijwani & Vadlani, 2011; Salgado et al.,
2015) or non productive adsorption to lignin hydrolyzates (Gao et al., 2014). On the contrary, Xyl
activity presented a dramatic decrease to almost zero value at 72h of fermentation and increased to
9±2 U g-1 DM by the end of the fermentation.
CMCase profile resulted in a completely different trend. Maximum CMCase activity was obtained at
the end of the fermentation, with a value of 5.67±0.43 U g-1 DM. Also, the second highest peak was
obtained at 16h of fermentation, just prior to maximum FPase, Xyl and BGase production, reaching a
value of 4.19±0.05 U g-1 DM. CMCase activity is related to the random degradation of the amorphous
sites of the cellulose chain, i.e, they are mainly responsible for the reduction of the polymerization
degree of cellulosic substrates to provide a more available substrate for exoglucanase action (Kostylev
& Wilson, 2012). Considering metabolic economics seems likely that CMCase would be produced
only when required for further hydrolysis.
The measurements for cellulase and xylanase activity are currently performed using several different
methods, some of them with many concerns on its reproducibility (Coward-Kelly et al., 2003). It is
Chapter 6. Reproducibility assessment
123
for this reason that the comparison between researches is complex; however, a summary of several
enzyme production processes are presented in Table 6.1.
Table 6.1. Summary of maximum cellulase and xylanase activities production (FPase, CMCase, BGase and Xyl) obtained by SSF.n.r: not reported, m.r: mixed inoculum. SH: soybean hulls, RH: rice husk, OP: orange peels, WS: wheat straw, RS: Rice straw, WB: wheat bran, POW: palm oil waste, SB: soybean, SCB sugarcane bagasse and CH: coffee husk.
Enzymatic activity (U g-1DM)
Substrate Amount
(g)
Inoculum SSF time
(h)
FPase CMCase
BGase Xyl References
SH:WB 100 Aspergillus oryzae
96 10.8 100.7 10.7 504.9 (Brijwani & Vadlani,
2011)
RH 5 Aspergillus oryzae
96 3.1 14.1 10.6 n.r (Bansal et al., 2012)
OP 5 Aspergillus oryzae
96 1.9 1.0 3.0 n.r (Bansal et al., 2012
WS 5 Aspergillus níger
96 13.6 11.2 5.5 n.r (Bansal et al., 2012
RS 10 Aspergillus níger
96 9.0 30.6 21.2 936.1 (Dhillon et al., 2011)
WB 10 Aspergillus níger
96 13.6 48.2 21.7 2601 (Dhillon et al., 2011)
WS 25 Trichoderma reesei
240 1.2 n.r n.r n.r (Lever et al., 2010)
POW 1 Aspergilllus turbingensis
120 2.4 n.r n.r 11.8 (Lever et al., 2010)
WB:SB 5 Aspergillis fumigatus
120 5.0 56 105.8 10.6 (Delabona et al., 2013)
SCB 5 Pleurotus ostreatus
72-120 0.25 0.13 n.r 11.0 (Membrillo et al., 2011)
CH 1,200 m.i 48 8-9 n.r n.r n.r Chapter 5
CH 15,200 m.i 24 3.08 2.51 0.13 44.51 This work
All enzymatic activities are produced in a wide range of values, probably due to the variety of inocula
and substrates used in those researches. FPase, CMCase and BGase production in this work was found
Chapter 6. Reproducibility assessment
124
in the reported range of production, as seen in Table 6.1. As for xylanase production, the reported
range of production (10.6 to 2601 U g-1 DM, according to Table 6.1) is wider than the reported for
cellulase. In spite of the low production values obtained in this work when comparing to literature, it
has to be stated that some authors have found even lower levels of enzymatic activities production
(Mansour et al., 2016).
Considering cellulase and xylanase production, it is remarkable the fact that most of the production
systems shown in Table 6.1 are carried out using small amounts of substrates, using pure strains and
long fermentation times. In this sense, the results obtained in this work are promising, due to the
significant cellulase and hemicellulase production in a short fermentation time, therefore allowing the
faster valorization of organic wastes and improving process economics.
Regarding fiber degradation, results are presented in Figure 6.3. Cellulose, hemicellulose and lignin
degradation started early with a 1.1, 12.4 and 7.0% degradation respectively in the first 8 h of
operation. During the most active stage of the process at 24 h of operation, partial degradations of
cellulose, hemicellulose and lignin were of 4.9, 13.4 and 4.1%, which are in accordance with reported
literature (Salgado et al., 2015). At the end of the fermentation a final cellulose degradation was
24.1%. In addition, final lignin hydrolysis was 11.25%, which is higher than the observed by other
researches in short solid-state fermentations (Umasaravanan et al., 2011). The most interesting result
was obtained for hemicellulose hydrolysis. A final degradation of 34.9% was achieved in spite of the
relatively low Xyl values as compared with literature (Table 6.1).
This process was performed using a non-sterile substrate, which provides a complex dynamic process,
involving different metabolisms. For this reason, enzymatic (or not enzymatic) products cannot be
properly correlated to operational parameters, but only expressed as the net result of these different
metabolisms. This might be the reason for not finding correlations among cellulases enzymatic
complex and xylanase activities production with their respective substrates. Even more, in all cases,
no correlation has been found using any of the parameters followed during the process. This is
probably due to the fact that parameters like sOUR or temperature, even though useful, provide an
extremely simplified overview of the process. For this reason, there is the need to seek for different
parameters or techniques to properly correlate the enzymatic production profile. In this sense, the
identification of microbial communities appears as an attractive alternative that enable the potential
direct correlation among substrate consumption and enzyme production.
Chapter 6. Reproducibility assessment
125
Figure 6.3. Cellulose, hemicellulose and lignin degradation profile obtained during pilot scale solid-
state fermentation using a specialized inoculum and coffee husk as substrate, during 134 h.
Degradation is expressed as mass percentage obtained from a mass balance.
c) Gaseous emissions
Figure 6.4 presents the evolution of the cumulative emissions of CH4, N2O, VOCs and NH3
throughout the SSF process carried our during the first pilot batch (R1).
Ammonia was emitted in higher amounts than the rest of the gases analyzed, and was mainly released
after the thermophilic stage at the highest biological activity condition as reported by other authors
(Maulini-Duran et al., 2015). Ammonia emission factor was 3.5·10-2 kg Mg-1 CH, obtained at 134 h.
Ammonia emission is mainly related to nitrogen compounds content such as ammonium from
proteins, which can be stripped out as ammonia through the exhausted gas outlet. VOCs, on the other
hand, were also found as one of the main contributors of emissions throughout the process with a total
emission factor of 1.1·10-2 kg Mg-1 CH at 134 h of fermentation. In spite of this value, when it is
considered a productive process for cellulase and xylanase, the SSF should take only 24 h, where the
emission factor was 2.6·10-3 kg Mg-1 CH.
Time [h]
0 20 40 60 80 100 120 140
Fibe
r deg
rada
tion
[%]
0
10
20
30
40
Lignin CelluloseHemicellulose
Chapter 6. Reproducibility assessment
126
Previous studies reported VOC emission factors 1000-fold higher than the obtained in this work
(Maulini-Duran et al., 2015), using orange peels (rich in limonene and other volatile compounds) as
substrate for cellulase production in a 5 day fermentation. VOCs emitted during the current process
presented an almost linear trend, indicating a nearly continuous degradation of the carbonaceous
material as suggested in other studies (Pagans et al., 2006b).
Figure 6.4. Cummulative gaseous emissions generated during pilot scale solid-state fermentation
using a specialized inoculum and coffee husk as substrate.
In contrast to NH3 and VOC emissions, CH4 and N2O are not directly related to composition but to the
appearance of anaerobic zones in the solid matrix of the reactor, especially at larger scales than those
reported in SSF literature. In the case of, N2O emissions, these are reported to be associated either to
anaerobic or anoxic conditions by means of the heterotrophic denitrification and nitrifier
denitrification processes during ammonia degradation as reported by some authors (Wunderlin et al.,
2013). In this study, CH4 and N2O emissions were only observed in very low amounts mainly during
the most active stage of the fermentation. In this context, the entire 134 h process presented total
cummulative emission factors of 9.4·10-4 and 2.6·10-4 kg Mg-1 CH for CH4 and N2O, respectively.
Time [h]
0 20 40 60 80 100 120 140
CH
4 [K
g M
g-1 C
H]
N2O
[Kg
Mg-1
CH
]
0
2x10-4
4x10-4
6x10-4
8x10-4
10-3
NH
3 [K
g M
g-1C
H]
VO
C [K
g M
g--1
CH
]
0
10-2
2x10-2
3x10-2
4x10-2
CH4N2O NH3VOC
Chapter 6. Reproducibility assessment
127
Even more, when considering the 24 h productive process, these factors even lower, achieving values
of 4.9·10-4 and 1.1·10-4 kg Mg-1 CH for CH4 and N2O, respectively. Other studies presented emissions
of these pollutants obtained during the highest biological activity in similar processes (Maulini-Duran
et al., 2015), which are in agreement with this work. However, the obtained emission factors were 10-
fold higher than those observed in that process. The absence of published data on the emissions of
pollutant gases from SSF unable the comparison of the obtained values, although in any case they are
lower than those of similar processes, such as composting (Pagans et al., 2006a; Pagans et al., 2006b).
The complete emission profile was performed in order to fully assess the potential environmental
impact and its correlation with enzyme production. However, if it is considered the period of time
when maximum enzymatic production is achieved, only VOC emissions are relevant due to NH3 it is
only produced after 32h of fermentation. This is of great importance, when considering a potential gas
treatment.
The operational measurements presented in Figures 6.2, 6.3 and 6.4 were performed in order to fully
understand the process dynamics and environmental impact, however, in terms of cellulase and
xylanase production, the process should be stopped at 24 h.
d) Replicates and consistency
In Figure 6.5a it is presented the first 24 h of operation of the previous SSF (see Figure 2 for the
complete profile) where maximum cellulase and xylanase activities were determined. Two replicates
of this 24 h fermentation are presented in Figure 5b and 5c.
Average of the initial pH and C/N ratio for the triplicates were 8.32±0.01 and 13.32±0.46
respectively. Initial temperature in the three pilot reactors were 27.8, 27.8 and 29.2ºC for
fermentations R1, R2 and R3, respectively. The optimal initial conditions of the inoculum along with
these temperature values allowed a rapid start-up of the fermentation, obtaining a very similar profile
during the first 8h. It is during this period that the thermophilic stage started in all replicates with an
average temperature of 47-48ºC. After this period differences on the sOUR profile appeared in R1
(Figure 5a) in comparison with R2 and R3. It is possible that the initial conditions of the inoculum
could have affected the performance of the pilot fermentations. Final sOUR and COC obtained at 24 h
of fermentation for the three processes were 3.6 mg O2 g-1 DM h-1 and 77.6 mg O2 g-1 DM for R1, 3.9
mg O2 g-1 DM h-1 and 65.5 mg O2 g-1 DM for R2 and 4.9 mg O2 g-1 DM h-1 and 71.8 mg O2 g-1 DM for
R3, respectively.
Chapter 6. Reproducibility assessment
128
Figure 6.5. Operation profile of OUR and temperature of the three replicates of the pilot solid-state
fermentation R1, R2 and R3 (50L) using a propagated specialized inoculum with coffee husk as
substrate.
Time [h]
sOU
R [m
gO2
g-1 D
M h
-1]
0
2
4
6
8
10
12
Tem
pera
ture
[ºC]
0
20
40
60
80
sOUR Temperature
sOU
R [m
gO2
g-1 D
M h
-1]
0
2
4
6
8
10
12
Tem
pera
ture
[ºC]
0
20
40
60
80
Time [h]
0 4 8 12 16 20
sOU
R [m
gO2
g-1 D
M h
-1]
0
2
4
6
8
10
12
Tem
pera
ture
[ºC]
0
20
40
60
80
R1
R2
R3
Chapter 6. Reproducibility assessment
129
Enzymatic activities obtained at 24 h of fermentation are presented in Table 2. ANOVA analysis
showed that only BGase production presented no significant differences among the three
fermentations. This enzyme is crucial during cellulose degradation, due to its role on cellobiose
degradation. BGase relieves the cellulase system for cellobiose accumulation and therefore, endo- and
exo-glucanases inhibition. BGase controls endo and exoglucanase action, which can act with a
synergetic effect, depending on the characteristics of the substrate (Kuhad et al., 2016; Rabinovich et
al., 2002).
FPase activity obtained in R2 and R3 presented no significant differences, as well as Xyl activity for
R1 and R3. Differences among the enzymatic yields can be attributed to the complexity of the
substrate and the nature of cellulolytic enzymes. Endoglucanases, reflected as CMCase, bound
randomly into any section of cellulose structure. This action directly affects all the following
hydrolysis steps with synergetic and repressive actions, which will affect the entire biodegradation
process (Kostylev & Wilson, 2012).
Additionally, in Table 6.2 is presented the average of enzymatic activities of the three fermentations
with their respective variation coefficient.
Table 6.2.Cellulase and xylanase yields obtained in the three replicates of the pilot solid-state
fermentation. Values of each enzymatic activity that do not share a letter are significantly different.
n.m not measured.
Replicate FPase
(FPU g-1DM)
CMCase
(U g-1DM)
BGase
(U g-1DM)
Xyl
(U g-1DM)
R1 3.63±0.20(A) 1.38±0.14(C) 0.12±0.00(E) 42.71±2.62(F)
R2 2.86±0.25(B) (n.m) 0.12±0.00(E) (n.m)
R3 2.91±0.37(B) 3.62±0.72(D) 0.15±0.06(E) 47.71±4.53(F)
Average ± s.d 3.08±0.49 2.51±1.31 0.13±0.04 44.51±8.30
Deviation coefficient
(%)
16 52 27 19
Chapter 6. Reproducibility assessment
130
FPase, BGase and Xyl production were 3.08±0.49 FPU g-1 DM, 0.13±0.04 U g-1 DM and 44.51±8.3 U
g-1 DM with a variation coefficient of 16, 27 and 19%, respectively. These variation coefficients can
be considered relatively high; however, when working with organic wastes the situation is different.
Variation coefficients only from proximal composition of the solid wastes can range between 10 to
50% (Leroy et al., 1992), therefore it is likely that any process that includes solid wastes has an
intrinsic high variation. Moreover, using mixed enzymatic compounds obtained by SSF in highly
controlled processes can achieve variation coefficients of nearly 10% (Martínez-Ruiz et al., 2008). In
this sense, the variation coefficients obtained in this work can be considered as acceptable, taking into
account that it is a complex process using non sterile material in a pilot scale.
As for CMCase, the average production was 2.51±1.31 U g-1 DM with a variation coefficient of 52%.
The high variation of the production of CMCase can be expected due to their random action into any
section of the cellulose structure and the impossibility to control its action, as mentioned before.
In light of these results combined with the differences among propagation reactors, it seems of great
importance to regulate or control the initial temperature of the fermentations, in order to reduce
possible variations in the fermentations. Initial differences determine the performance of the pilot
fermentations with a major influence on the development of microbial communities.
e) Microbial characterization
Bacteria
A total of 30 bacterial families were identified in the final products of propagation reactors (P1, P2
and P3) and the pilot fermentations (R1, R2 and R3). The full composition is presented in Figure 6.6.
As seen in Figure 6, there are three main dominant families in all assessed samples: Paenibacillaceae,
Xanthomonadaceae and Sphingobacteriaceae; these are also the main families identified in the
original inoculum obtained in Chapter 5, Figure 5.5. The sum of these families account for a 45.6,
47.9 and 43.4% for P1, P2 and P3 and 40.5, 40.9 and 39.4% for R1, R2 and R3, respectively. In this
context, the addition of the specialized inoculum to a propagation reactor and then as inoculum for a
pilot SSF generated great similarities among the bacterial diversity at family level. Regarding the
samples obtained from the propagation reactors, P1 and P2 presented similar relative abundance of the
microbial populations; however, P3 showed slight differences. For instance, P1 and P2 showed low
presence of the family Flavobacteriaceae (2.1 and 2.4% for P1 and P2), while P3 showed significant
abundance this family (13.4%). The main specie found in this family was Flavobacterium
anatoliense, which is a strict aerobic bacteria isolated from several environments that has been
Chapter 6. Reproducibility assessment
131
reported as unable to grow on cellulose and, even more, it has been reported not to be able to
hydrolyze complex polysaccharides (Kacagan et al., 2013). In spite of the differences, the
predominant specie found in all propagation reactor samples was the same: Pseudoxanthomonas
taiwanensis. This bacterium was found predominant in the original specialized inoculum (Cerda et al.,
2017) and it was able to survive and colonize the propagation reactors, achieving a relative abundance
of 14.7, 14,9 and 9.6% in P1, P2 and P3 respectively. P. taiwanensis has been widely related to
cellulose degradation systems, for its β-glucosidase production and the potential enhancement of the
growth of other cellulolytic bacteria (Eichorst et al., 2013).
The fermented solids obtained from the propagation reactors were used as inoculum for the pilot SSF.
The final product of the latter was analyzed for bacteria identification, which results are presented in
Figure 6.6.
Figure 6.6. Bacterial distributions at the family level according to the16s sequencing for the final
products of propagation reactors (P1, P2 and P3) and pilot fermentations at highest cellulase activity
(R1, R2 and R3). Only families with a relative abundance >1% in all samples are depicted. Families
detected with a relative abundance < 1% in these samples are grouped as "others".
0 10 20 30 40 50 60 70 80 90 100
P1
P2
P3
R1
R2
R3
Relative abundance (%)
Paenibacillaceae Xanthomonadaceae Sphingobacteriaceae ChitinophagaceaeBacillaceae Trueperaceae Phyllobacteriaceae CytophagaceaeFlavobacteriaceae Alcaligenaceae ]Micrococcaceae ClostridiaceaeCorynebacteriaceae Thermoanaerobacteraceae Planococcaceae ComamonadaceaeAnaerolineaceae No Hit Thioalkalispiraceae Unclasified (Tissierellia class)Thermoactinomycetaceae Caldicoprobacteraceae Rhodanobacteraceae MicrobacteriaceaeRhodobacteraceae Pseudomonadaceae Moraxellaceae SymbiobacteriaceaeCaulobacteraceae Idiomarinaceae Unknown Others (<1%)
Chapter 6. Reproducibility assessment
132
The results showed that the same families predominant on propagation reactors were predominant on
the pilot SSF. In spite of the great resemblance, there is a shift on the predominant specie when
comparing with propagation reactor. In this sense, Sphingobacteriaceae family became more relevant
when compared with the propagation reactors, with a 27.9, 28.9 and 22.9% of relative abundance for
R1, R2 and R3, and 10.9, 12.2 and 18.8 for P1, P2 and P3, respectively. The harsh conditions
generated during pilot SSF could have affected the development of the different microorganisms in
the solid matrix. In this context, strong thermophillic and alkaline environments favored the growth of
bacteria able to thrive at these conditions, among them Sphingobacterium thermophillum (15.8, 11.1
and 9.7% for R1, R2 and R3) and Sphingobacterium arenae (3.6, 7.8 and 6.8% for R1, R2 and R3).
The new dominant species relegate P.taiwanensis to a specie with a minor presence in the solid
matrix, with a relative abundance of 2.4, 2.7 and 8.2% for R1, R2 and R3, respectively. S.
thermophilum and S. arenae have been reported to be present during the thermophillic stage of
composting processes with a potential for β-glucosidase production (Yabe et al., 2013). In addition, it
has been found that several cellulase obtained under these conditions can be halotolerant (Gladden et
al., 2014), which is of great interest when a it is considered to use in a potential use in bioethanol
production.
In general, the relative abundance at family level presented high similarities in propagation reactors
and the pilot SSF triplicates. Also, it is interesting to highlight the fact that families with relative
abundance below 1% (registered in Figure 6.6 as "others") are higher in the propagation reactors than
in the pilot SSF. This indicates a further reduction on biological diversity probably associated with the
pilot SSF conditions as confirmed by the Shannon diversity index for propagation reactors P1, P2 and
P3 were 2.6, 2.4 and 2.2 and for pilot SSF were 1.8, 1.7 and 1.9 for R1, R2 and R3 respectively.
Fungi and yeasts
A total number of 21 fungal families have been identified in the final products of propagation reactors
(P1, P2 and P3) and the pilot fermentations (R1, R2 and R3). The full composition is presented in
Figure 6.7. As seen in this figure, Phaffomycetaceae was found as the dominant family in all the
assessed samples. This family accounted for a total of 74.4, 64.5 and 48.2% for P1, P2 and P3, and
79.7, 61.9 and 77.8% for R1, R2 and R3, which represented most of the biological diversity of the
process. This family was also found as predominant in the initial inoculum, with a relative abundance
of 43.2% (Chapter 5, Figure 5.6).
Chapter 6. Reproducibility assessment
133
Figure 6.7. Fungal distributions at the family level according to the ITS sequencing for the final
products of propagation reactors (P1, P2 and P3) and pilot fermentations at highest cellulase activity
(R1, R2 and R3). Only families with a relative abundance >1% in all samples are depicted. Families
detected with a relative abundance < 1% in these samples are grouped as "others".
These results proved, at family level, a consistent propagation and colonization of the specialized
mycobiota present in the initial inoculum. Even more, when performing the pilot SSF, the biological
diversity was further reduced, maintaining Phaffomycetaceae as the predominant family and
increasing its relative abundance. Two species of this family appear as dominant in propagation
reactors: Cyberlidnera jardinii with 64.3, 55.1 and 3.5% for P1, P2 and P3 and Barnettozyma
californica with a 10.1, 9.4 and 43.6% for P1, P2 and P3 respectively. These two species were also
found as dominant in the initial inoculum (Chapter 5, Figure 5.6) with a relative abundance of 34 and
9.1% for C.jardinii and B. californica, respectively. There is a clear difference on the relative
abundance of these species in P3 when compared with P1 and P2. Both of these species are related to
hemicellulose degradation with potential tolerance to grow in the presence of lignin hydrolyzates,
which is of great interest for a potential use on bioethanol production (Morais et al., 2013; Nordberg
et al., 2014).
0 10 20 30 40 50 60 70 80 90 100
P1
P2
P3
R1
R2
R3
Relative abundance (%)
Phaffomycetaceae Ascosphaeraceae No HitHarpochytriaceae Monoblepharidaceae Unclassified (Polyporales order)Unknown 1 Unknown 2 Unknown 3Unknown 4 Unknown 5 Unknown 6Unknown 7 Unknown 8 Unknown 9Unknown 10 Unknown 11 Unknown 12Unknown 13 Unknown 14 Unknown 15Others (<1%)
Chapter 6. Reproducibility assessment
134
Identified families from the fermented solids from the pilot SSF are presented in Figure 6.7. These
results presented great similarities, at family level, with propagation reactors, although a major shift
occurred at specie level. In pilot SSF the dominant specie was B. californica with a 72.6, 59.7 and
72.8% of relative abundance, leaving C. jardinii as a minor component of the mycobiota with an
abundance below 3% in all cases. B. californica, as mentioned before, is related to hemicellulose
degradation and an important xylanase producer; however, it has been also reported as an enzyme
producer for lignin degradation (Martorell et al., 2012). This specie is a producer of tyrosinase, which
is an enzyme able to oxidize phenolic compounds that can potentially enhance lignin hydrolizates
degradation. The great abundance of this specie in the pilot SSF can, in a way, explain the high
degradation of lignin in the reactors (above 11% of degradation).
In general, in the pilot SSF, most of the identified mycobiota presented high hemicellulose and lignin
degradation potential. Improvements of degradation on these structures can make cellulose structure
more accessible for the enzymes to act and, therefore, increase cellulase production. In addition,
biological diversity obtained in the pilot reactors was reduced when compared with the propagation
reactors. Shannon index for R1, R2 and R2 were 0.44, 0.62 and 0.50 and 1.02, 0.99 and 0.97 for P1,
P2 and P3, respectively.
The most important result on these experiments relies on the fact that when the operational conditions
of the reactors are similar, the biomass that is able to colonize the reactors is the same, even when the
fermentation is carried out using non sterile substrates. On the contrary, when the scale or operational
conditions such as temperature changed, there is a shift in microbial communities affecting slightly
cellulase and xylanase production.
As a final remark, it has to be stated that there is no previous research performed on the
reproducibility of pilot scale SSF. Many aspects considered as SSF drawbacks have been undertaken,
resulting in the development of a more robust and sustainable process. The use of a standard inoculum
using wastes as substrate makes this technology more consistent for its application at larger scales,
with the additional economical benefit of using organic wastes as raw material.
6.4. Conclusion
The use of a specialized inoculum on a pilot scale SSF presented very promising results. Cellulase and
xylanase production at pilot scale was successful, achieving values in accordance with the literature in
a short period of time. The gaseous emissions detected showed that the process generates low
amounts of pollutants, below the reported emissions in SSF or similar processes. Only VOCs were
Chapter 6. Reproducibility assessment
135
emitted in significant amounts, obtaining almost negligible emission values of NH3, N2O and CH4.
The triplicates carried out for the pilot SSF presented small differences in enzymatic production;
however, in general, the values remained in the expected range of production. The strong conditions
of temperature and pH provided in the pilot SSF affected the productive process and also the
developing of the microbial diversity. Biological diversity reduction, reflected on Shannon index, was
observed for both bacteria and fungi when the process was scaled up. Considering the possibility of a
more efficient waste management with the production of a value-added product, the proposed process
can be described as a sustainable, robust and environmentally friendly process.
References
Abraham, J., Gea, T., Sánchez, A. 2013. Potential of the solid-state fermentation of soy fibre residues by native microbial populations for bench-scale alkaline protease production. Biochemical Engineering Journal, 74, 15-19.
Bansal, N., Tewari, R., Soni, R., Soni, S. 2012. Production of cellulases from Aspergillus niger NS-2 in solid state fermentation on agricultural and kitchen waste residues. Waste Management, 32(7), 1341-1346.
Brijwani, K., Vadlani, P. 2011. Cellulolytic Enzymes Production via Solid-State Fermentation: Effect of Pretreatment Methods on Physicochemical Characteristics of Substrate. Enzyme Research, 2011, 1-10.
Cerda, A., El-Bakry, M., Gea, T., Sánchez, A. 2016. Long term enhanced solid-state fermentation: Inoculation strategies for amylase production from soy and bread wastes by Thermomyces sp. in a sequential batch operation. Journal of Environmental Chemical Engineering, 4(2), 2394-2401.
Cerda, A., Gea, T., Vargas-García, M., Sánchez, A. 2017. Towards a competitive solid state fermentation: cellulases production from coffee husk by sequential batch operation and role of microbial diversity. Science of the Total Environment.
Coward-Kelly, G., Aiello-Mazzari, C., Kim, S., Granda, C., Holtzapple, M. 2003. Suggested improvements to the standard filter paper assay used to measure cellulase activity. Biotechnology and Bioengineering, 82(6), 745-749.
Delabona, P.d.S., Pirota, R.D.P.B., Codima, C.A., Tremacoldi, C.R., Rodrigues, A., Farinas, C.S. 2013. Effect of initial moisture content on two Amazon rainforest Aspergillus strains cultivated on agro-industrial residues: Biomass-degrading enzymes production and characterization. Industrial Crops and Products, 42, 236-242.
Dhillon, G., Oberoi, H., Kaur, S., Bansal, S., Brar, S. 2011. Value-addition of agricultural wastes for augmented cellulase and xylanase production through solid-state tray fermentation employing mixed-culture of fungi. Industrial Crops and Products, 34(1), 1160-1167.
Eichorst, S., Varanasi, P., Stavila, V., Zemla, M., Auer, M., Singh, S., Simmons, B., Singer, S. 2013. Community dynamics of cellulose-adapted thermophilic bacterial consortia. Environmental Microbiology, 15(9), 2573-2587.
Gao, D., Haarmeyer, C., Balan, V., Whitehead, T., Dale, B., Chundawat, S. 2014. Lignin triggers irreversible cellulase loss during pretreated lignocellulosic biomass saccharification. Biotechnology for Biofuels, 7(1), 175.
Gladden, J., Park, J., Bergmann, J., Reyes-Ortiz, V., D’haeseleer, P., Quirino, B., Sale, K., Simmons, B., Singer, S. 2014. Discovery and characterization of ionic liquid-tolerant thermophilic
Chapter 6. Reproducibility assessment
136
cellulases from a switchgrass-adapted microbial community. Biotechnology for Biofuels, 7(1), 15.
Kacagan, M., Inan, K., Belduz, A., Canakci, S. 2013. Flavobacterium anatoliense sp. nov., isolated from fresh water, and emended description of Flavobacterium ceti. International Journal of Systematic and Evolotionary Microbiology, 63(Pt 6), 2075-81.
Kostylev, M., Wilson, D. 2012. Synergistic interactions in cellulose hydrolysis. Biofuels, 3(1), 61-70. Kuhad, R., Deswal, D., Sharma, S., Bhattacharya, A., Jain, K., Kaur, A., Pletschke, B., Singh, A.,
Karp, M. 2016. Revisiting cellulase production and redefining current strategies based on major challenges. Renewable and Sustainable Energy Reviews, 55, 249-272.
Leroy, D., Giovannoni, J., Maystre, L. 1992. Sampling method to determine a household waste composition variance. Waste Management & Research, 10(1), 3-12.
Lever, M., Ho, G., Cord-Ruwisch, R. 2010. Ethanol from lignocellulose using crude unprocessed cellulase from solid-state fermentation. Bioresource Technology, 101(18), 7083-7087.
Mansour, A., Arnaud, T., Lu-Chau, T., Fdz-Polanco, M., Moreira, M., Rivero, J. 2016. Review of solid state fermentation for lignocellulolytic enzyme production: challenges for environmental applications. Reviews in Environmental Science and Bio/Technology, 15(1), 31-46.
Martínez-Ruiz, A., García, H., Saucedo-Castañeda, G., Favela-Torres, E. 2008. Organic Phase Synthesis of Ethyl Oleate Using Lipases Produced by Solid-state Fermentation. Applied Biochemistry and Biotechnology, 151(2), 393-401.
Martorell, M., Pajot, H., de Figueroa, L. 2012. Dye-decolourizing yeasts isolated from Las Yungas rainforest. Dye assimilation and removal used as selection criteria. International Biodeterioration & Biodegradation, 66(1), 25-32.
Maulini-Duran, C., Abraham, J., Rodríguez-Pérez, S., Cerda, A., Jiménez-Peñalver, P., Gea, T., Barrena, R., Artola, A., Font, X., Sánchez, A. 2015. Gaseous emissions during the solid state fermentation of different wastes for enzyme production at pilot scale. Bioresource Technology, 179, 211-218.
Membrillo, I., Sánchez, C., Meneses, M., Favela, E., Loera, O. 2011. Particle geometry affects differentially substrate composition and enzyme profiles by Pleurotus ostreatus growing on sugar cane bagasse. Bioresource Technology, 102(2), 1581-1586.
Morais, C., Cadete, R., Uetanabaro, A., Rosa, L., Lachance, M., Rosa, C. 2013. D-xylose-fermenting and xylanase-producing yeast species from rotting wood of two Atlantic Rainforest habitats in Brazil. Fungal Genetics and Biology, 60, 19-28.
Nordberg, H., Cantor, M., Dusheyko, S., Hua, S., Poliakov, A., Shabalov, I., Smirnova, T., Grigoriev, I., Dubchak, I. 2014. The genome portal of the Department of Energy Joint Genome Institute: 2014 updates. Nucleic Acids Research, 42, 26-31.
Pagans, E., Barrena, R., Font, X., Sánchez, A. 2006a. Ammonia emissions from the composting of different organic wastes. Dependency on process temperature. Chemosphere, 62(9), 1534-1542.
Pagans, E., Font, X., Sánchez, A. 2006b. Emission of volatile organic compounds from composting of different solid wastes: Abatement by biofiltration. Journal of Hazardous Materials, 131(1-3), 179-186.
Puyuelo, B., Gea, T., Sánchez, A. 2010. A new control strategy for the composting process based on the oxygen uptake rate. Chemical Engineering Journal, 165(1), 161-169.
Rabinovich, M., Melnick, M., Bolobova, A. 2002. The Structure and Mechanism of Action of Cellulolytic Enzymes. Biochemistry (Moscow), 67(8), 850-871.
Ruggieri, L., Gea, T., Artola, A., Sánchez, A. 2009. Air filled porosity measurements by air pycnometry in the composting process: A review and a correlation analysis. Bioresource Technology, 100(10), 2655-2666.
Salgado, J., Abrunhosa, L., Venâncio, A., Domínguez, J., Belo, I. 2015. Enhancing the bioconversion of winery and olive mill waste mixtures into lignocellulolytic enzymes and animal feed by Aspergillus uvarum using a packed-bed bioreactor. Journal of Agricultural and Food Chemistry, 63(42), 9306-9314.
Chapter 6. Reproducibility assessment
137
Sundh, I., Rönn, S. 2002. Microbial Succession During Composting of Source-Separated Urban Organic Household Waste Under Different Initial Temperature Conditions. in: Microbiology of Composting, (Eds.) H. Insam, N. Riddech, S. Klammer, Springer Berlin Heidelberg. Berlin, Heidelberg, pp. 53-64.
Umasaravanan, D., Jayapriya, J., Rajendran, R. 2011. Comparison of lignocellulose biodegradation in solid state fermentation of sugarcane bagasse and rice straw by Aspergillus tamarii. Ceylon Journal of Science (Biological Sciences), 40(1), 65-68.
Wunderlin, P., Lehmann, M., Siegrist, H., Tuzson, B., Joss, A., Emmenegger, L., Mohn, J. 2013. Isotope Signatures of N2O in a Mixed Microbial Population System: Constraints on N2O Producing Pathways in Wastewater Treatment. Environmental Science & Technology, 47(3), 1339-1348.
Yabe, S., Aiba, Y., Sakai, Y., Hazaka, M., Kawahara, K., Yokota, A. 2013. Sphingobacterium thermophilum sp. nov., of the phylum Bacteroidetes, isolated from compost. International Journal of Systematic and Evolutionary Microbiology, 63(Pt 5), 1584-8.
Chapter 7. Long-term amylase production
137
Chapter 7. Long-term amylase production
138
CHAPTER 7. INOCULATION STRATEGIES FOR
AMYLASE PRODUCTION IN A SEQUENTIAL
BATCH OPERATION
This chapter has been published in Journal of Environmental Chemical Engineering. "Long term
enhanced solid-state fermentation: Inoculation strategies for amylase production from soy and bread
wastes by Thermomyces sp. in a sequential batch operation".
Cerda, A., El-Bakry, M., Gea, T., Sánchez, A.
Chapter 7. Long-term amylase production
139
Chapter 7. Long-term amylase production
140
Summary
In parallel to the experiments for cellulase production explained in detailed in chapter 5, other
experiments were carried out for the production of amylase using soy fiber and bread waste as the
substrate. Amylase production was assessed due to the fact that during this period of time a
postdoctoral researcher: Dr. Mamdouh El-Bakry, was carrying out lab scale SSF using thermophilic
bacteria as inoculum, with positive results. In light of the success on preeliminar tests using 4.5 L
reactors, it was decided to develop different reinoculation strategies, in order to reach a continuous
operation. The full description of the experimental methodology and the obtained results are detailed
in the following section. In addition, a brief introduction to amylase production "state of the art" is
presented.
7.1 Reinoculation strategies to enhance amylase production in solid-state fermentation using
Thermomyces sp. as inoculum.
7.1.1 Background
In the last decade solid-state fermentation (SSF) has received increased attention from researchers as a
bio-based process. This technology allows for the valorisation of wastes for the production of valuable
commodities such as enzymes (Saxena et al., 2011; El-Bakry et al., 2015; Pitol et al., 2016).
SSF presents some advantages in comparison with conventional fermentations such as higher
productivity or reduced energy requirements, and low wastewater output (Pitol et al., 2016). However,
important drawbacks are described concerning the SSF scale up largely due to heat transfer,
sterilization costs and culture heterogeneity (Mitchell et al., 2006; Casciatori et al., 2016). Abraham et
al. (2013) proposed to perform SSF as a thermophilic batch process under near-adiabatic conditions,
similar to the composting process which is a robust, well known, easily scalable and low-cost process.
This self-heating process does not involve heating/cooling costs since the temperature evolve due to
the heat released in the biodegradation process. Based on this proposal, El-Bakry et al. (2016) proved
that thermophilic strains inoculated on non-sterile wastes were able to grow and compete with
autochthonous microbiota and significantly increase the protease production. Still the setting up of a
long term operation in continuous or semi-continuous regime remains a challenge and very few
attempts have been reported (Alkasrawi et al., 2013; Cheirsilp and Kitcha, 2015; Farinas, 2015).
Enzymatic products such as amylases have great potential in different biotechnological applications
such as food, fermentation, detergent, textile and paper industries (Farinas, 2015). Some thermophilic
amylase-producers microorganisms have been found during the highest bio-oxidative activity stage
Chapter 7. Long-term amylase production
141
during composting processes(Zeng et al., 2005; López-Gonzalez et al., 2015a; López-Gonzalez et al.,
2015b). In addition, it has been well established that these microorganisms are able to produce
substantial amounts of amylase in SSF systems (Babu and Satyanarayana, 2005; Kunamneni et al.,
2005; Saxena et al., 2011).
Several attempts for the optimization of enzymatic production have been reported, such as nutritional
supplementation for microorganism growth (Petrova et al., 2000; Mukherjee et al., 2009; Hashemi et
al., 2013), use of engineered strains (Petrova et al., 2000; Rajagopalan and Krishna, 2010) or
operational strategies such as batch or fed-batch (Tao et al., 1998; Astolfi et al., 2011; Cheirsilp and
Kitcha, 2015). However, most SSF for the production of enzymatic compounds have been tested at
lab scale and under sterile substrate conditions (Farinas, 2015).
The first aim of this work was to develop a SSF process for the production of amylases using a
thermophilic strain growing on non-sterile wastes in the lab scale (500-mL reactors) and bring it to the
bench scale (10-L reactors), based on the proposal of Abraham et al. (2013) and El-Bakry et al.
(2016). Second goal was to perform a first approach to the assessment of the effect of inoculation
strategies in sequential batch operation in order to develop a long term operation process in a solid-
state fermentation configuration at the bench scale. SSF of organic wastes operated in sequential
batches was evaluated for the production of amylases using thermophilic and amylase producer
microorganisms (Thermomyces sp. and Geobacillus sp.).
7.1.2 Materials
Soy fibre residues (SF) were provided by Soy Nature (Barcelona, Spain). Bread waste (BW, expired
pre-cooked baguettes) was procured from a local market. Both wastes were obtained in a sufficient
amount to do the entire experiments with the same material as substrate. Materials were used as
received in 500 mL reactors and in the case of the inoculation strategies experiments, aliquots were
frozen separately prior to their use. Characterization of both substrates is presented in Table 7.1.
Moisture was initially adjusted to 70% using tap water in all the experiments.
In these experiments, the inoculum was selected among two thermophilic and amylase producers:
Geobacillus sp. ATCC-31198 and Thermomyces sp. ATCC-200065. These microorganisms were
incubated in Petri dishes on supplier´s recommended agar media: ATCC® Media 1207 and ATCC®
Media 350 respectively.
Chapter 7. Long-term amylase production
142
Table 7.1. Characterization of raw materials used on SSF (average ± standard deviation) * all
percentages in dry basis except for dry mass, in wet basis.
Parameter* Soy fiber (SF) Bread waste (BW)
Dry mass (%) 17.7±1.3 62.8±0.8
Organic matter (%) 96.8±1.3 96.95±0.2
C (%) 67.5±1.8 48.1±2.1
N (%) 4.4±0.7 2.5±0.2
pH 7.0±0.7 6.1±0.1
Starch (%) 3.1±0.1 75.7±0.1
Both strains were grown in a liquid medium containing the same formulation than that used for Petri
dishes. Final pH was adjusted to 7.0 ± 0.2 and 100 mL were incubated in a 500 mL Erlenmeyer flask
at 150 rpm and 55ºC for 16h. Final culture broth was used as inoculum in all lab scale fermentations
and in the propagation reactor for sequential batch experiments in a 10% w/w ratio.
7.1.3 Experimental procedure
a) Selection of inoculum type and size.
Solid-state fermentations were carried out at 55°C for 7 days in 500 mL Erlenmeyer flasks with 100 g
of wet substrate and at fixed airflow of 20 mL min-1. Also, 10 g of wood sticks were added to the
substrate as a bulking agent to provide porosity. Oxygen content was measured continuously in the
reactors gas output. Specific oxygen uptake rate (sOUR) was calculated with equation 1. These
experiments were performed using the set up described in chapter 3, section 3.1.4
Samples were taken daily for amylase activity determination. Firstly, Geobacillus sp. and
Thermomyces sp. were evaluated in the solid-state fermentation of SF and BW for the production of
amylases and the best inoculum was selected for further experiments. Secondly, mixtures of different
SF:BW ratios (0:100, 10:90, 50:50, 90:10 and 100:0 (w/w) were evaluated and the best mixture
selected for following experiments in reactors. Experiments were carried out in triplicates.
Chapter 7. Long-term amylase production
143
b) Scaling-up of amylase production
The scaling-up effect on the process defined at lab scale was assessed by monitoring two 4.5 L
reactors (described in chapter 3, section 3.1.5) for 14 days in order to have a complete profile of all
relevant parameters of fermentation. One reactor was operating without inoculation as control. A
second reactor was inoculated directly with liquid Thermomyces sp. culture. Amylase activity was
determined at days 0, 4, 7, 10 and 14 of operation.
c) Inoculation strategies for operation in sequential batches
Experiments were performed using the same system that in section 7.1.3.b.
One first batch was performed as a propagation reactor (PR) and it was inoculated directly with
Thermomyces sp. liquid culture in a 10% w/w ratio according to El-Bakry et al. (2016). Samples were
taken from this reactor in two different moments to act as inoculum for the following batches. Three
sequential batches (SB) were performed and each batch was inoculated with solids from the previous
batch. Inoculation size was 10% w/w (fermented solids / fresh solids in the reactor) in all cases. The
three strategies assayed are described below:
• Strategy MOUR: Using as inoculum the fermented solids obtained at the moment of
maximum sOUR, that is, solids with Thermomyces sp. at maximum growth rate.
• Strategy MAA: Using as inoculum the fermented solids obtained at the end of the process,
that is, solids obtained at the moment of maximum amylase activity (AA).
• Strategy MAAE: Using as inoculum the solids obtained after amylase recovery when its
maximum activity was reached at the end of the process.
A scheme of the inoculation strategies is presented in Figure 7.1
Sampling was performed on a composite sample of the reactor, after homogenizing all the reactor
material at the time of substrate exchange. These samples were analyze on its amylase activity
production and starch content. In addition, routine analysis described in chapter 3, section 3.3 were
performed to the solid samples.
Chapter 7. Long-term amylase production
144
Figure 7.1. Scheme of the reinoculation strategies developed to enhance amylase production.
7.1.4 Results
a) Selection of inoculum type and size.
Fig. 7.2 shows sOUR and amylase activity profiles for experiments in 500 mL reactors for inoculum
selection using SF (Fig. 7.2a and b) and BW (Fig. 7.2and d) as substrate. As observed in Fig. 7.2a and
1c, higher sOUR values were obtained in reactors inoculated with Thermomyces sp. and Geobacillus
sp. strains than in non-inoculated reactors, when using SF and BW as sole substrate. This increase was
more obvious in the case of soy fibre, which presented a higher sOUR than bread waste confirming a
higher biodegradability. sOUR values were maximum at around 24h in all the cases and slowly
decreased after that. Similar results were obtained for amylase activity: inoculation showed a positive
effect on amylase production for both substrates, and higher enzymatic yields were obtained for SF
than for BW. Maximum activity was obtained 48 h after reaching the maximum value of sOUR and
slightly decreased after that. For both substrates, reactors inoculated with Thermomyces sp. presented
higher values of sOUR and AA than reactors with Geobacillus sp. Best process performance was
obtained using SF and Thermomyces sp. with a maximum sOUR of 11.8 mgO2 g-1DM h-1 and a
Propagationreactor
SB1 SB3SB2
SB3SB2SB1
SB3SB2SB1Extraction
Maximum sOUR10% of the total substrate
Maximum AA10% of the total substrate
10% 10%
10% 10%
10% 10%
Enzyme rich-liquidfraction
Chapter 7. Long-term amylase production
145
maximum amylase production of 34·103 U g-1DS. Fungi were expected to be better inoculum due to
their ability to produce all required enzymes to fully degrade polymeric compounds and also to be
able to survive in extreme conditions (De Gannes et al., 2013). Also, fungi play a key role in solid
fermentations such as composting processes, where Thermomyces sp. has been detected during
thermophilic and highest amylolytic activity production stage (Zhang et al., 2014; López-Gonzalez et
al., 2015). These authors reported the predominance of Thermomyces sp. in temperatures ranging 50-
60ºC and alkaline pH, which is also the range of its optimum growth regarding temperature and pH.
Figure 7.2. Solid-state fermentation at 500 mL scale using Thermomyces sp. and Geobacillus sp. as
inoculums. Profiles of sOUR and amylase activity are presented using soy fiber a) and b) and bread
wastes c) and d) respectively.
In consequence, Thermomyces sp. was selected as the best strain. De Castro et al (2013) reported the
synergistic effect of different wastes used as substrates for amylase production by SSF. Accordingly,
further experiments were performed with different mixtures of SF and BW to assess this potential
synergistic effect. Table 7.2 summarizes maximum sOUR and amylase activity for all evaluated
mixtures using different ratios of SF:BW. It was found that sOUR and amylase production were
improved at high SF content due to its higher content and availability of easily biodegradable organic
matter. Also the mineral content of SF is higher than BW content, especially in Mg, P, and K which
sOU
R [m
gO2
g-1D
M h
-1]
0
2
4
6
8
10
12
Time [d]0 1 2 3 4 5
sOU
R [m
gO2
g-1D
M h
-1]
0
2
4
6
8
10
12No inoculation Geobacillus Thermomyces
Am
ylas
e ac
tivity
[U g
-1D
S]
0
10000
20000
30000
40000
Time [d]0 1 2 3 4 5
Am
ylas
e ac
tivity
[U g
-1D
S]0
10000
20000
30000
40000No inoculation GeobacillusThermomyces
a b
c d
Chapter 7. Long-term amylase production
146
are required for the growth of amylase producing microorganisms (Tuncel et al., 2014; Kim et al.,
2015). Starch contained on BW most probably acted as inducer for amylase production as reported by
some researchers (Tao et al., 1998; Mukherjee et al., 2009) and nutrients present on SF provided
proper conditions for the proliferation of microorganisms. It has been reported that several cations
present in great amounts on SF such as Ca2+, acted as stabilizing agent or co-factors for amylases
(Petrova et al., 2003; Mukherjee et al., 2009), generating an amylase activity increase, which is in
accordance with obtained results.
All the mixtures showed a maximum amylase production on day 4 of process. Maximum amylase
activity was obtained at a 90:10 SF:BW ratio, with a value over 39.9 ·103 U g-1DS and maximum
sOUR of 11.5 mgO2 g-1DM h-1. Lowest amylase production was obtained with BW as sole substrate,
with a maximum activity of 10.9·103 U g-1DS. As it is possible to observe in all enzymatic profiles,
amylase production is not directly related to maximum biological activity or growth. This fact was
expected as amylase production is regulated by starch inducible system and repressed in presence of
glucose or mannose isomers (Song, 1990). Maximum amylase was observed 48h after maximum
sOUR as also reported by Kunamneni et al. (2005) and Petrova et al. (2003).
Table 7.2. Maximum sOUR and amylase activity obtained in SSF using different SF:BW ratios.
SF:BW (%w/w) Maximum sOUR
(mgO2 g-1DM h-1)
Maximum amylase activity
(103 U g-1DS)
0:100 3.4 10.9 ± 0.5
10:90 4.1 14.7 ± 0.7
50:50 5.8 18.9 ± 0.9
90:10 11.5 39.9 ± 0.9
100:00 11 34.6 ± 1.2
Chapter 7. Long-term amylase production
147
b) Scale-up effect on amylase production.
The selected process (substrates mixture 90:10 w/w and Thermomyces sp) was evaluated in 4.5 L
reactors and results are presented in Figure 7.3. Maximum sOUR and temperature were obtained at
day 2 of operation in both control and inoculated reactors. As observed at lab scale, maximum
biological activity did not match with amylase maximum production, which was obtained 48h after
the maximum sOUR (day 4) in both cases. There was a clear effect of inoculation in terms of
biological activity and amylase activity production (El-Bakry et al., 2016). Thermomyces sp.
inoculated reactor presented a temperature of 65ºC and a maximum sOUR of 11.1 mgO2 g-1DM h-1
which represented an 8 and 47% increase respectively in comparison with non-inoculated reactors.
Also, using Thermomyces sp. as inoculum presented a positive effect on amylase activity with a 36%
increment in comparison with control reactor, reaching a maximum of 41.03·103 U g-1DS.
Similar results were obtained at 4.5 L and 500 mL scales for maximum sOUR and amylase activity.
This confirms the easy scalability of the self-heating SSF and its suitability to work with thermophilic
strains with non-sterile substrates.
Figure 7.3. Solid-state fermentation at 4.5 L scale using SF:BW 90:10 ratio as substrate. Profiles of
sOUR, temperature and amylase activity are presented using a) no inoculation, b) Themomyces sp.
inoculation.
Time [d]
0 2 4 6 8 10 12 14
Tem
pera
ture
[°C
]
sOU
R [m
gO2-1
g-1 D
M h
-1]
0
10
20
30
40
50
60
70
Time [d]
0 2 4 6 8 10 12 14
Am
ylas
e ac
tivity
[U g
-1D
S]
0
10000
20000
30000
40000
50000
60000
Temperature sOUR Amylase activity
a. No inoculation b. Thermomyces sp. inoculation
Chapter 7. Long-term amylase production
148
c) Inoculation strategies for operation in sequential batches.
Process profile for the propagation reactor is presented in Figure. 7.4a, 7.5a and 7.6a. A lag phase of
nearly 2 days was observed. This was due to the use of previously frozen substrates. As reported by
Pognani et al. (2012), freezing for less than one year does not affect the biodegradable organic matter
content or the biological activity measured as sOUR, but a lag phase may appear. Thermophilic
conditions were reached after that and remained nearly for 48h, achieving maximum sOUR of 23
mgO2 g-1DM h-1 after sampling the reactor. Sampling sOUR was 17 mgO2 g-1DM h-1 and obtained
amylase activity of 31.7·103 U g-1DS. Total oxygen consumed was 725 mgO2 g-1DM in 5 days. At the
end of the fermentation (48h after maximum sOUR) amylase activity was 15% higher with a value of
36.6·103 U g-1DS.
• Strategy MOUR
Figure 7.4 presents the process profiles for PR plus three sequential batches (SB) performed with
inoculation using solids obtained at the moment of maximum sOUR (gap in temperature curve of SB2
was due to data acquisition failure). In all four cases, maximum sOUR was achieved after sampling,
which could be of importance considering the quality of the inoculums for the sequential batches (El-
Bakry et al., 2016).
The three sequential batches presented maximum temperatures ranging between 53-61°C similar to
PR. As for sOUR profile, maximum values were obtained also after sampling, reaching values of 21,
23, 19 and 19 mgO2 g-1DM h-1 for PR and SB 1, 2 and 3 respectively (Figure 7.4).
Amylase activity obtained during maximum sOUR were 31.7, 21.9, 7.1 and 43.0·103 U g-1DS and
during maximum AA 36.6, 39.1, 47.2 and 29.3· 103 U g-1DS for PR and SB1, 2 and 3 respectively.
In general terms, sOUR increase and lag phase reduction was only obtained during SB 1. Latter
inoculations resulted in lower or equal sOUR, obtained only after homogenization of reactors in the
sampling process. In addition, each sequential inoculation produced an increase of 0.8, 0.9 and 15.1%
in total oxygen consumption in comparison with propagation reactor.
Chapter 7. Long-term amylase production
149
Figure 7.4. MOUR strategy in 4.5 L reactors using SF:BW 90:10 ratio as substrate. Profiles of
sOUR, temperature and amylase activity are presented for PR and three sequential batches MOUR-
SB1, MOUR-SB2 and MOUR-SB3.
Inoculation with solids from maximum sOUR period from PR was expected to provide a reduction in
lag phase and an increase in sOUR. This is based on the fact that fungi from the class Eurotiomycetes,
such as Thermomyces sp. have been found predominant in thermophillic stage during composting
processes (Ugwuanyi et al., 2004; López-Gonzalez et al., 2015a; López-Gonzalez et al., 2015b),
indicating that a suitable environment for inoculated Thermomyces sp. is provided. This fact was
observed through all sequential batches, where a slight increase in sOUR and total oxygen
consumption was obtained.
• Strategies MAA and MAAE
Figure 7.5 presents process profiles for PR and SB 1, 2 and 3 for MAA strategy. Temperature profile
was similar in all four processes, reaching values near 60°C. Maximum sOUR values were in the
range of 21-25 mgO2 g-1DM h-1 except for the first sequential batch (Figure 7.5b). However, total
oxygen consumption presented a sustained increase of 1.9, 54.2 and 76.6% in the SB 1, 2 and 3
respectively in comparison with the PR. Also lag phase was gradually reduced in each SB, and sOUR
peaked in the mesophilic phase in SB 2 and 3 with values 8-9 mgO2 g-1DM h-1.
Am
ylas
e ac
tivity
[U g
-1D
S]
0
10000
20000
30000
40000
50000
Time [d]0 1 2 3 4 5
Tem
pera
ture
[°C
]sO
UR
[mgO
2 g-1
DM
h-1
]
0
10
20
30
40
50
60
70
Time [d]0 1 2 3 4 5
Am
ylas
e ac
tivity
[U g
-1D
S]
0
10000
20000
30000
40000
50000
Tem
pera
ture
[°C
]sO
UR
[mgO
2 g-
1 DM
h-1
]
0
10
20
30
40
50
60
70
Temperature sOUR Amylase activity
Propagation reactor MOUR-SB1
MOUR-SB2 MOUR-SB3
Chapter 7. Long-term amylase production
150
During thermophilic period (temperature >45ºC), high amylase activity was produced in all batches.
Highest amylase production of 228.9·103 U g-1DS was obtained at the end of the MAA-SB2, which
represents a 500% increase in comparison with the propagation reactor. In the MAA-SB3, amylase
activity was 156.9·103 U g-1DS, 329% higher than PR.
Figure 7.5. MAA strategy in 4.5 L reactors using SF:BW 90:10 ratio as substrate. Profiles of sOUR,
temperature and amylase activity are presented for PR and three sequential batches MAA-SB1, MAA-
SB2 and MAA-SB3.
Figure 7.6 shows the process profile for PR and the three sequential batches, using final solids after
the extraction of soluble and enzymatic components as inoculums. This solid material would be the
actual final output of the process after amylase recovery as targeted product.
Maximum sOUR in the three SB were lower than in the PR. A 16.6 % decrease in the total oxygen
consumption on first SB was obtained but increased by 1.1 and 23.2 % in SB 2 and SB 3.
Thermophilic phase was longer in the three SB than in PR. Amylase activity was higher in the three
SB than in PR, with an increase of 27, 98, 56% in SB 1, 2 and 3 respectively. Maximum activity of
the sequential experiments was obtained at the end of batch 2, achieving a value of 72.4·103 U g-1DS.
Time [d]
0 2 4 6 8 10 12 14 16 18
Tem
pera
ture
[°C
]sO
UR
[mgO
2-1g-
1 DM
h-1
]
0
10
20
30
40
50
60
70
Am
ylas
e A
ctiv
ity [U
g-1
DS]
0
50000
100000
150000
200000
250000
Temperature sOUR Amylase activity
Propagationreactor
MAA-SB1 MAA-SB2 MAA-SB3
Chapter 7. Long-term amylase production
151
Figure 7.6. MAAE strategy in 4.5 L reactors using SF:BW 90:10 ratio as substrate. Profiles of sOUR,
temperature and amylase activity are presented for PR and three sequential batches MAAE-SB1,
MAAE-SB2 and MAAE-SB3.
Regarding MAA and MAAE strategies, temperature profiles and total oxygen consumption presented
a different trend than PR. Temperature is known as a crucial parameter to describe microbial activity
during aerobic fermentation of solid heterogeneous material. It has been stated that inoculation with
different microorganisms to a solid state fermentation could lead to a rapid rate of temperature
elevation and for a more prolonged time of high-temperature process (Jurado et al., 2015).
In MAA strategy, inoculation with fermented solid inoculum with high enzymatic and soluble content
in sequential batches may have generated a fast hydrolysis of available substrate for the consumption
of all present microorganisms. On the other hand, an extraction of soluble and enzymatic compounds
in the inoculums could have generated a reduction in sOUR values and in the oxygen consumption
rate, due to a lack of hydrolytic capacity in comparison with propagation reactor and the other
evaluated batches with different inoculation strategies.
Time [d]
0 2 4 6 8 10 12 14 16 18 20 22
Tem
pera
ture
[°C
]sO
UR
[mgO
2-1g-
1 DM
h-1
]
0
10
20
30
40
50
60
70
Am
ylas
e A
ctiv
ity [U
g-1
DS]
0
50000
100000
150000
200000
250000
Temperature sOUR Amylase activity
Propagationreactor
MAAE-SB1 MAAE-SB2 MAAE-SB3
Chapter 7. Long-term amylase production
152
A reduction in amylase activity in the SB3 in both MAA and MAAE strategies can be attributed to
several factors such as the possible repression on enzymatic synthesis (Babu and Satyanarayana.,
1995; Mukherjee et al., 2009), presence of hydrolytic enzymes such as proteases (Abraham et al.,
2013), to a depletion of the amorphous form of the solid substrate (Tao et al., 1998) or even more, to a
change in the production profile of amylases.
It is a fact that sugars and low polymerization degree polysaccharides are normally fast consumed by
microflora present during fermentation. It is reasonable to assume that with high production of
amylase activity, high amounts of sugars are released which could cause catabolic repression of
amylase synthesis. On the other hand, researchers have presented that catabolic repression at high
reducing sugars content occurs even in metabolically engineered de-repressed strains (Rajagopalan
and Krishnan, 2010).
The lack of literature about amylase production by inoculation strategies in sequential solid state
fermentations and the many different methods of amylase activity determination made difficult to
compare the obtained results with other reported researches. However, other enzymes have been
produced in fed batch strategy (Cheirsilp and Kitcha, 2015), with results found in accordance to this
study. In any case it is proved that the operation strategy is a key factor for the implementation of
solid state fermentation at the commercial level. Further research shall focus on determining the
number of sequential batches that can be performed following this strategy. Nevertheless, both MAA
and MAAE strategies allowed a remarkable enhancement on amylase activity. In our opinion, both
strategies are suitable for industrial operation: in strategy MAA, 10% of the solids would be diverted
to inoculate the following batch prior to enzyme extraction; in strategy MAAE, 10% of the solids after
extraction would be recycled to inoculate the next batch. To finally decide which the best strategy is,
it would be necessary to perform a complete economical and environmental assessment.
7.1.4 Conclusion
Operation of SSF of soy and bread wastes in sequential batches has been proven as a suitable strategy
for amylase production with Thermomyces sp. Additionally, it opens a new valorisation alternative for
these wastes. A SSF process was operated for 19 days and the enhancement of amylase production
was accomplished. In terms of amylase yield, the most suitable inoculation strategy was using
fermented solids at the end of the batch to inoculate the following batch, obtaining a 500% increase in
productivity and eliminating the need of producing fresh inoculum for each batch. Nonetheless this,
the final strategy selection should consider economical and environmental aspects. The development
Chapter 7. Long-term amylase production
153
of these operational strategies could benefit process economics and provides the possibility of a
continuous operation for amylases and potentially for other enzymes.
References
El-Bakry, M., Abraham, J., Cerda, A., Barrena, R., Ponsá, S., Gea, T., Sánchez, A. 2015. From Wastes to High Value Added Products: Novel Aspects of SSF in the Production of Enzymes. Critical Reviews on Environmental Science and Technology, 45, 1999-2042.
Saxena, R., Singh, R. 2011. Amylase production by solid-state fermentation of agro-industrial wastes using Bacillus sp. Brazilian Journal of Microbiology, 42, 1334-1342.
Pitol, L., Biz, A., Mallmann, E., Krieger, N., Mitchell, D. 2016. Production of pectinases by solid-state fermentation in a pilot-scale packed-bed bioreactor. Chemical Engineering Journal, 283, 1009-1018.
Mitchell, D., Berovič, M., Krieger, N. 2006. Solid-State Fermentation Bioreactor Fundamentals: Introduction and Overview. in: Solid-State Fermentation Bioreactors, (Eds.) D. Mitchell, M. Berovič, N. Krieger, Springer Berlin Heidelberg, pp. 1-12.
Abraham, J., Gea, T., Sánchez, A. 2013. Potential of the solid-state fermentation of soy fibre residues by native microbial populations for bench-scale alkaline protease production. Biochemical Engineering Journal, 74, 15-19.
Astolfi, V., Joris, J., Verlindo, R., Oliveira, J.V., Maugeri, F., Mazutti, M., de Oliveira, D., Treichel, H. 2011. Operation of a fixed-bed bioreactor in batch and fed-batch modes for production of inulinase by solid-state fermentation. Biochemical Engineering Journal, 58–59, 39-49.
Babu, K., Satyanarayana, T. 1995. α-Amylase production by thermophilic Bacillus coagulans in solid state fermentation. Process Biochemistry, 30, 305-309.
Casciatori, F., Bück, A., Thoméo, J., Tsotsas, E. 2016. Two-phase and two-dimensional model describing heat and water transfer during solid-state fermentation within a packed-bed bioreactor. Chemical Engineering Journal, 287, 103-116.
Cheirsilp, B., Kitcha, S. 2015. Solid state fermentation by cellulolytic oleaginous fungi for direct conversion of lignocellulosic biomass into lipids: Fed-batch and repeated-batch fermentations. Industrial Crops and Products, 66, 73-80.
De Castro, R., Sato, H. 2013. Synergistic effects of agroindustrial wastes on simultaneous production of protease and α -amylase under solid state fermentation using a simplex centroid mixture design. Industrial Crops and Products, 49, 813-821.
De Gannes, V., Eudoxie, G., Hickey, W. 2013. Insights into fungal communities in composts revealed by 454-pyrosequencing: implications for human health and safety. Frontiers Microbiology, 4, 164.
El-Bakry, M., Gea, T., Sánchez, A. 2016. Inoculation effect of thermophilic microorganisms on protease production through solid-state fermentation under non-sterile conditions at lab and bench scale (SSF). Bioprocess and Biosystem Engineering, 39(4),585-592.
Farinas, C. 2015. Developments in solid-state fermentation for the production of biomass-degrading enzymes for the bioenergy sector. Renewable and Sustainable Energy Reviews, 52, 179–188.
Hashemi, M., Mousavi, S., Razavi, S., Shojaosadati, S. 2013. Comparison of submerged and solid state fermentation systems effects on the catalytic activity of Bacillus sp. KR-8104 α-amylase at different pH and temperatures. Industrial Crops and Products, 43, 661-667.
Kim, H., Lee, Y., Kim, Y. 2015. Efficacy of pectin and insoluble fiber extracted from soy hulls as a functional non-meat ingredient. Food Science and Technology, 64(2), 1071-1077.
Jurado, M., Suárez-Estrella, F., López, M., Vargas-García, M., López-González, J., Moreno, J. 2015. Enhanced turnover of organic matter fractions by microbial stimulation during lignocellulosic waste composting. Bioresource Technology, 186, 15-24.
Chapter 7. Long-term amylase production
154
Kunamneni, A., Permaul, K., Singh, S. 2005. Amylase production in solid state fermentation by the thermophilic fungus Thermomyces lanuginosus. Journal of Bioscience and Bioengineering, 100, 168-171.
López-González, J., Suárez-Estrella, F., Vargas-García, M., López, M., Jurado, M., Moreno, J. 2015a. Dynamics of bacterial microbiota during lignocellulosic waste composting: Studies upon its structure, functionality and biodiversity. Bioresource Technology, 175, 406-416.
López-González, J., Vargas-García, M., López, M., Suárez-Estrella, F., Jurado, M., Moreno, J. 2015b. Biodiversity and succession of mycobiota associated to agricultural lignocellulosic waste-based composting. Bioresource Technology, 187, 305-313.
Mukherjee, A., Borah, M., Rai, S. 2009. To study the influence of different components of fermentable substrates on induction of extracellular α -amylase synthesis by Bacillus subtilis DM-03 in solid-state fermentation and exploration of feasibility for inclusion of α-amylase in laundry detergent formulations. Biochemical Engineering Journal, 43, 149-156.
Petrova, S., Ilieva, S., Bakalova, N., Atev, A., Bhat, M., Kolev, D. 2000. Production and characterization of extracellular α -amylases from the thermophilic fungus Thermomyces lanuginosus (wild and mutant strains). Biotechnology Letters, 22, 1619-1624.
Pognani, M., Barrena, R., Font, X., Sánchez, A. 2012. Effect of freezing on the conservation of the biological activity of organic solid wastes. Bioresource Technology, 104, 832-836.
Rajagopalan, G., Krishnan, C. 2010. Hyper-production of alpha-amylase from agro-residual medium with high-glucose in SSF using catabolite derepressed Bacillus subtilis KCC103. Journal of Basic Microbiology, 50, 336-43.
Song, M. 1990. Regulation of Alpha-amylase Synthesis in Bacillus Subtilis. University of Wisconsin-Madison.
Swain, M., Kar, S., Padmaja, G., Ray, R. 2006. Partial characterization and optimization of production of extracellular α -amylase from Bacillus subtilis isolated from culturable cow dung microflora. Polish Journal of Microbiology, 55, 289-296.
Tao, S., Beihui, L., Peng, L., Deming, L., Zuohu, L. 1998. New solid-state fermentation process for repeated batch production of fibrinolytic enzyme by Fusarium oxysporum. Process Biochemistry, 33, 419-422.
The US Department of Agriculture and The US Composting Council, Test methods for the examination of composting and compost, Edaphos International, Houston, 2001.
Tuncel, N. Yılmaz, H. Kocabıyık, A. Uygur, A. 2014. The effect of infrared stabilized rice bran substitution on B vitamins, minerals and phytic acid content of pan breads: Part II. Journal of Cereal Science, 59(2), 162–166.
Ugwuanyi, J., Harvey, L., McNeil, B. 2004. Development of thermophilic populations, amylase and cellulase enzyme activities during thermophilic aerobic digestion of model agricultural waste slurry. Process Biochemistry, 39, 1661-1669.
Zeng, G., Yu, M., Chen, Y., Huang, D., Zhang, J., Huang, H., Jiang, R., Yu, Z. 2010. Effects of inoculation with Phanerochaete chrysosporium at various time points on enzyme activities during agricultural waste composting. Bioresource Technology , 101, 222-227.
Zhang, X., Zhong, Y., Yang, S., Zhang, W., Xu, M., Ma, A., Zhuang, G., Chen, G., Liu, W. 2014. Diversity and dynamics of the microbial community on decomposing wheat straw during mushroom compost production. Bioresource Technology, 170, 183-195.
Appendix
155
Appendix
156
APPENDIX
Appendix
157
Appendix
158
Appendix I. Calibration curve for FPase determination.
Figure I.
Reducing sugars [g L-1]
0 2 4 6
Abs
orba
ncia
[nm
]
0.0
0.5
1.0
1.5
2.0
ABS=0.297*[RS]R2= 0.992
Appendix
159
Appendix II. Calibration curve for CMCase determination.
Figure II.
Reducing sugars [g L -1]
0.0 0.5 1.0 1.5 2.0
Abs
orba
nce
[nm
]
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
ABS= 0.310*[RS]
R2=0.996
Appendix
160
Appendix III. Comparative between substrates in cellulase determination.
In order to partially assess the difficulties to work with filter paper as a substrate for cellulase
determination, microcrystalline cellulose was evaluated as a potential substrate in all experiments
carried out in chapters 5 and 6. Figure III shows the correlation obtained using all the experimental
data.
There is no evident correlation among the two substrates, which indicates that the effect of the
enzymatic pool obtained in each fermentation affects differently the two substrates. In first place, it
can be observed that on one hand, cellulase activity using microcrystalline cellulose (SC) increases
quickly and reaches a plateau, while using filter paper as the substrate (FP) the increment is gradual.
In second place, the obtained values of cellulose activity using SC substrate were substantially higher
than using filter paper. Maximum cellulase activity using microcrystalline cellulose as substrate
resulted in values 2 or even 3-fold higher than when filter paper was used as substrate.
This difference can be somehow explained by considering the following aspects: heterogeneity of
insoluble cellulose, unclear dynamic interactions between insoluble substrate and cellulase
components, and the complex competitive and/or synergic relationships among all three enzymes that
SC= 30.1*(1-exp(-0.6*FP)) -7.4R2=0.42
FP [FPU g-1DM]
0 2 4 6 8 10 12
SC [U
g-1
DM
]
0
5
10
15
20
25
30
35
Appendix
161
are involved in overall cellulase effects. All these aspects could limit the setup of proper strategies,
depending on activity screening approaches.
According to some authors, SC could become a preferred substrate to replace filter paper because it is
a more recalcitrant substrate, yielding a more stringent substrate for total cellulase activity than filter
paper does; and activities measured on microcrystalline cellulose could more accurately represent
hydrolysis ability on pretreated lignocellulose, because its characteristics are closer to those of
pretreated lignocelluloses, based on cellulose accessibility to cellulase and the degree of
polymerization (Zhang and Lynd, 2004).
Microcrystalline cellulose has two characteristics that could explain the differences of cellulose
activity measurement with two different substrates. In first place, microcrystalline cellulose has
almost one third of the glycosidic bonds content of filter paper. This is important because
endocellulase acts upon these bonds, which would imply a better performance using filter paper as
substrate. However, another important characteristic of microcrystalline cellulose corresponds to its
polymerization degree. This parameter is very low (150-500) in microcrystalline cellulose in
comparison with filter paper (100-2800), which could imply a higher accessibility of exocellulases for
the product released for the action of endocellulase (Percival Zhang et al., 2006). In this sense, even if
endocellulases could perform better on filter paper, exocellulases are reported to be responsible for
most of the product released.
In addition, it is important to take into account that sugar quantification by DNS considers reducing
sugars (with reducing ends), which implies several considerations. One of the most important
parameter is the reducing ends content in the different substrates, which for SC and FP are 0.67 and
0.13% respectively (Percival Zhang et al., 2006). In this sense, hydrolysis of microcrystalline
cellulose can release much more reducing ends than filter paper as in these experiments, but not
necessarily would consider total cellulose activity, which could be the possible explanation for the
differences obtained by using two different substrates.
Furthermore, in both cases, the results may suffer from an underestimation of cellulase activity when
glucose is used as the standard and β-glucosidase is not in excess. Due to the relatively low BGase
values obtained in all cases, it is highly possible that both activity measurements are underestimated.
Appendix
162
Appendix
163